Daily Papers

We show that when large language models learn to reward hack on production RL environments, this can result in egregious emergent misalignment. We start with a pretrained model, impart knowledge of reward hacking strategies via synthetic document finetuning or prompting, and train on a selection of real Anthropic production coding environments. Unsurprisingly, the model learns to reward hack. Surprisingly, the model generalizes to alignment faking, cooperation with malicious actors, reasoning about malicious goals, and attempting sabotage when used with Claude Code, including in the codebase for this paper. Applying RLHF safety training using standard chat-like prompts results in aligned behavior on chat-like evaluations, but misalignment persists on agentic tasks. Three mitigations are effective: (i) preventing the model from reward hacking; (ii) increasing the diversity of RLHF safety training; and (iii) "inoculation prompting", wherein framing reward hacking as acceptable behavior during training removes misaligned generalization even when reward hacking is learned.

Recent studies reveal that integrating new modalities into Large Language Models (LLMs), such as Vision-Language Models (VLMs), creates a new attack surface that bypasses existing safety training techniques like Supervised Fine-tuning (SFT) and Reinforcement Learning with Human Feedback (RLHF). While further SFT and RLHF-based safety training can be conducted in multi-modal settings, collecting multi-modal training datasets poses a significant challenge. Inspired by the structural design of recent multi-modal models, where, regardless of the combination of input modalities, all inputs are ultimately fused into the language space, we aim to explore whether unlearning solely in the textual domain can be effective for cross-modality safety alignment. Our evaluation across six datasets empirically demonstrates the transferability -- textual unlearning in VLMs significantly reduces the Attack Success Rate (ASR) to less than 8\% and in some cases, even as low as nearly 2\% for both text-based and vision-text-based attacks, alongside preserving the utility. Moreover, our experiments show that unlearning with a multi-modal dataset offers no potential benefits but incurs significantly increased computational demands, possibly up to 6 times higher.

Safety and security remain critical concerns in AI deployment. Despite safety training through reinforcement learning with human feedback (RLHF) [ 32], language models remain vulnerable to jailbreak attacks that bypass safety guardrails. Universal jailbreaks - prefixes that can circumvent alignment for any payload - are particularly concerning. We show empirically that jailbreak detection systems face distribution shift, with detectors trained at one point in time performing poorly against newer exploits. To study this problem, we release JailbreaksOverTime, a comprehensive dataset of timestamped real user interactions containing both benign requests and jailbreak attempts collected over 10 months. We propose a two-pronged method for defenders to detect new jailbreaks and continuously update their detectors. First, we show how to use continuous learning to detect jailbreaks and adapt rapidly to new emerging jailbreaks. While detectors trained at a single point in time eventually fail due to drift, we find that universal jailbreaks evolve slowly enough for self-training to be effective. Retraining our detection model weekly using its own labels - with no new human labels - reduces the false negative rate from 4% to 0.3% at a false positive rate of 0.1%. Second, we introduce an unsupervised active monitoring approach to identify novel jailbreaks. Rather than classifying inputs directly, we recognize jailbreaks by their behavior, specifically, their ability to trigger models to respond to known-harmful prompts. This approach has a higher false negative rate (4.1%) than supervised methods, but it successfully identified some out-of-distribution attacks that were missed by the continuous learning approach.

Large language models (LLMs) show remarkable human-like capability in various domains and languages. However, a notable quality gap arises in low-resource languages, e.g., Indonesian indigenous languages, rendering them ineffective and inefficient in such linguistic contexts. To bridge this quality gap, we introduce Cendol, a collection of Indonesian LLMs encompassing both decoder-only and encoder-decoder architectures across a range of model sizes. We highlight Cendol's effectiveness across a diverse array of tasks, attaining 20% improvement, and demonstrate its capability to generalize to unseen tasks and indigenous languages of Indonesia. Furthermore, Cendol models showcase improved human favorability despite their limitations in capturing indigenous knowledge and cultural values in Indonesia. In addition, we discuss the shortcomings of parameter-efficient tunings, such as LoRA, for language adaptation. Alternatively, we propose the usage of vocabulary adaptation to enhance efficiency. Lastly, we evaluate the safety of Cendol and showcase that safety in pre-training in one language such as English is transferable to low-resource languages, such as Indonesian, even without RLHF and safety fine-tuning.

In this work, we introduce the PKU-SafeRLHF dataset, designed to promote research on safety alignment in large language models (LLMs). As a sibling project to SafeRLHF and BeaverTails, we separate annotations of helpfulness and harmlessness for question-answering pairs, providing distinct perspectives on these coupled attributes. Overall, we provide 44.6k refined prompts and 265k question-answer pairs with safety meta-labels for 19 harm categories and three severity levels ranging from minor to severe, with answers generated by Llama-family models. Based on this, we collected 166.8k preference data, including dual-preference (helpfulness and harmlessness decoupled) and single-preference data (trade-off the helpfulness and harmlessness from scratch), respectively. Using the large-scale annotation data, we further train severity-sensitive moderation for the risk control of LLMs and safety-centric RLHF algorithms for the safety alignment of LLMs. We believe this dataset will be a valuable resource for the community, aiding in the safe deployment of LLMs.

Current methods for content safety in Large Language Models (LLMs), such as Supervised Fine-Tuning (SFT) and Reinforcement Learning from Human Feedback (RLHF), often rely on multi-stage training pipelines and lack fine-grained, post-deployment controllability. To address these limitations, we propose a unified co-training framework that efficiently integrates multiple safety behaviors: positive (lawful/prosocial), negative (unfiltered/risk-prone) and rejective (refusal-oriented/conservative) within a single SFT stage. Notably, each behavior is dynamically activated via a simple system-level instruction, or magic token, enabling stealthy and efficient behavioral switching at inference time. This flexibility supports diverse deployment scenarios, such as positive for safe user interaction, negative for internal red-teaming, and rejective for context-aware refusals triggered by upstream moderation signals. This co-training strategy induces a distinct Safety Alignment Margin in the output space, characterized by well-separated response distributions corresponding to each safety mode. The existence of this margin provides empirical evidence for the model's safety robustness and enables unprecedented fine-grained control. Experiments show that our method matches the safety alignment quality of SFT+DPO, with our 8B model notably surpassing DeepSeek-R1 (671B) in safety performance, while significantly reducing both training complexity and deployment costs. This work presents a scalable, efficient, and highly controllable solution for LLM content safety.

Despite notable advancements in Multimodal Large Language Models (MLLMs), most state-of-the-art models have not undergone thorough alignment with human preferences. This gap exists because current alignment research has primarily achieved progress in specific areas (e.g., hallucination reduction), while the broader question of whether aligning models with human preferences can systematically enhance MLLM capability remains largely unexplored. To this end, we introduce MM-RLHF, a dataset containing 120k fine-grained, human-annotated preference comparison pairs. This dataset represents a substantial advancement over existing resources, offering superior size, diversity, annotation granularity, and quality. Leveraging this dataset, we propose several key innovations to improve both the quality of reward models and the efficiency of alignment algorithms. Notably, we introduce a Critique-Based Reward Model, which generates critiques of model outputs before assigning scores, offering enhanced interpretability and more informative feedback compared to traditional scalar reward mechanisms. Additionally, we propose Dynamic Reward Scaling, a method that adjusts the loss weight of each sample according to the reward signal, thereby optimizing the use of high-quality comparison pairs. Our approach is rigorously evaluated across 10 distinct dimensions and 27 benchmarks, with results demonstrating significant and consistent improvements in model performance. Specifically, fine-tuning LLaVA-ov-7B with MM-RLHF and our alignment algorithm leads to a 19.5% increase in conversational abilities and a 60% improvement in safety. We have open-sourced the preference dataset, reward model, training and evaluation code, as well as reward modeling and safety benchmarks. For more details, please visit our project page: https://mm-rlhf.github.io.

Existing approaches to language model alignment often treat safety as a tradeoff against helpfulness, which can lead to unacceptable responses in sensitive domains. To ensure reliable performance in such settings, we propose High-Confidence Safe Reinforcement Learning from Human Feedback (HC-RLHF), a method that provides high-confidence safety guarantees while maximizing helpfulness. Similar to previous methods, HC-RLHF explicitly decouples human preferences into helpfulness and harmlessness (safety), which are learned by training a reward model and a cost model, respectively. It then employs a two-step process to find safe solutions. In the first step, it optimizes the reward function under an intentionally pessimistic version of the cost constraint. In the second step, the trained model undergoes a safety test to verify whether its performance stays within an upper-confidence bound of the actual cost constraint. We provide a theoretical analysis of HC-RLHF, including proof that it will not return an unsafe solution with a probability greater than a user-specified threshold. For our empirical analysis, we apply HC-RLHF to align three different language models (Qwen2-1.5B, Qwen2.5-3B, and LLaMa3.2-3B) with human preferences. Our results demonstrate that HC-RLHF produces safe models with high probability and can improve harmlessness and helpfulness compared to previous methods.

With the widespread application of Large Language Models (LLMs), it has become a significant concern to ensure their safety and prevent harmful responses. While current safe-alignment methods based on instruction fine-tuning and Reinforcement Learning from Human Feedback (RLHF) can effectively reduce harmful responses from LLMs, they often require high-quality datasets and heavy computational overhead during model training. Another way to align language models is to modify the logit of tokens in model outputs without heavy training. Recent studies have shown that contrastive decoding can enhance the performance of language models by reducing the likelihood of confused tokens. However, these methods require the manual selection of contrastive models or instruction templates. To this end, we propose Adversarial Contrastive Decoding (ACD), an optimization-based framework to generate two opposite system prompts for prompt-based contrastive decoding. ACD only needs to apply a lightweight prompt tuning on a rather small anchor dataset (< 3 min for each model) without training the target model. Experiments conducted on extensive models and benchmarks demonstrate that the proposed method achieves much better safety performance than previous model training-free decoding methods without sacrificing its original generation ability.

Effective conversational agents must be able to personalize their behavior to suit a user's preferences, personality, and attributes, whether they are assisting with writing tasks or operating in domains like education or healthcare. Current training methods like Reinforcement Learning from Human Feedback (RLHF) prioritize helpfulness and safety but fall short in fostering truly empathetic, adaptive, and personalized interactions. Traditional approaches to personalization often rely on extensive user history, limiting their effectiveness for new or context-limited users. To overcome these limitations, we propose to incorporate an intrinsic motivation to improve the conversational agents's model of the user as an additional reward alongside multi-turn RLHF. This reward mechanism encourages the agent to actively elicit user traits by optimizing conversations to increase the accuracy of its user model. Consequently, the policy agent can deliver more personalized interactions through obtaining more information about the user. We applied our method both education and fitness settings, where LLMs teach concepts or recommend personalized strategies based on users' hidden learning style or lifestyle attributes. Using LLM-simulated users, our approach outperformed a multi-turn RLHF baseline in revealing information about the users' preferences, and adapting to them.

Large reasoning models (LRMs) "think" by generating structured chain-of-thought (CoT) before producing a final answer, yet they still lack the ability to reason critically about safety alignment and are easily biased when a flawed premise is injected into their thought process. We propose RECAP (Robust Safety Alignment via Counter-Aligned Prefilling), a principled reinforcement learning (RL) method for post-training that explicitly teaches models to override flawed reasoning trajectories and reroute to safe and helpful responses. RECAP trains on a mixture of synthetically generated counter-aligned CoT prefills and standard prompts, requires no additional training cost or modifications beyond vanilla reinforcement learning from human feedback (RLHF), and substantially improves safety and jailbreak robustness, reduces overrefusal, and preserves core reasoning capability -- all while maintaining inference token budget. Extensive analysis shows that RECAP-trained models engage in self-reflection more frequently and remain robust under adaptive attacks, preserving safety even after repeated attempts to override their reasoning.

Reinforcement learning from human feedback (RLHF) is a technique for training AI systems to align with human goals. RLHF has emerged as the central method used to finetune state-of-the-art large language models (LLMs). Despite this popularity, there has been relatively little public work systematizing its flaws. In this paper, we (1) survey open problems and fundamental limitations of RLHF and related methods; (2) overview techniques to understand, improve, and complement RLHF in practice; and (3) propose auditing and disclosure standards to improve societal oversight of RLHF systems. Our work emphasizes the limitations of RLHF and highlights the importance of a multi-faceted approach to the development of safer AI systems.

Reinforcement Learning with Human Feedback (RLHF) is a methodology designed to align Large Language Models (LLMs) with human preferences, playing an important role in LLMs alignment. Despite its advantages, RLHF relies on human annotators to rank the text, which can introduce potential security vulnerabilities if any adversarial annotator (i.e., attackers) manipulates the ranking score by up-ranking any malicious text to steer the LLM adversarially. To assess the red-teaming of RLHF against human preference data poisoning, we propose RankPoison, a poisoning attack method on candidates' selection of preference rank flipping to reach certain malicious behaviors (e.g., generating longer sequences, which can increase the computational cost). With poisoned dataset generated by RankPoison, we can perform poisoning attacks on LLMs to generate longer tokens without hurting the original safety alignment performance. Moreover, applying RankPoison, we also successfully implement a backdoor attack where LLMs can generate longer answers under questions with the trigger word. Our findings highlight critical security challenges in RLHF, underscoring the necessity for more robust alignment methods for LLMs.

Reinforcement learning from human feedback (RLHF) is a recent technique to improve the quality of the text generated by a language model, making it closer to what humans would generate. A core ingredient in RLHF's success in aligning and improving large language models (LLMs) is its reward model, trained using human feedback on model outputs. In machine translation (MT), where metrics trained from human annotations can readily be used as reward models, recent methods using minimum Bayes risk decoding and reranking have succeeded in improving the final quality of translation. In this study, we comprehensively explore and compare techniques for integrating quality metrics as reward models into the MT pipeline. This includes using the reward model for data filtering, during the training phase through RL, and at inference time by employing reranking techniques, and we assess the effects of combining these in a unified approach. Our experimental results, conducted across multiple translation tasks, underscore the crucial role of effective data filtering, based on estimated quality, in harnessing the full potential of RL in enhancing MT quality. Furthermore, our findings demonstrate the effectiveness of combining RL training with reranking techniques, showcasing substantial improvements in translation quality.

Reinforcement Learning from Human Feedback (RLHF) is a vital strategy for enhancing model safety in language models. However, annotating preference data for RLHF is a resource-intensive and creativity-demanding process, while automatic generation methods face limitations in data diversity and quality. In response, we present Safer-Instruct, a novel pipeline for semi-automatically constructing large-scale preference datasets. Our approach leverages reversed instruction tuning, instruction induction, and expert model evaluation to efficiently generate high-quality preference data without human annotators. We evaluate Safer-Instruct using LLaMA for instruction induction and GPT-4 as an expert model, generating approximately 10K preference samples. Finetuning an Alpaca model on this dataset demonstrates improved harmlessness while maintaining competitive performance on conversation and downstream tasks. Safer-Instruct addresses the challenges in preference data acquisition, advancing the development of safer and more responsible AI systems. Our code and data are available at https://github.com/uscnlp-lime/safer-instruct

Multimodal large language models (MLLMs) are critical for developing general-purpose AI assistants, yet they face growing safety risks. How can we ensure that MLLMs are safely aligned to prevent undesired behaviors such as discrimination, misinformation, or violations of ethical standards? In a further step, we need to explore how to fine-tune MLLMs to enhance reasoning performance while ensuring they satisfy safety constraints. Fundamentally, this can be formulated as a min-max optimization problem. In this study, we propose Safe RLHF-V, the first multimodal safety alignment framework that jointly optimizes helpfulness and safety using separate multimodal reward and cost models within a Lagrangian-based constrained optimization framework. Given that there is a lack of preference datasets that separate helpfulness and safety in multimodal scenarios, we introduce BeaverTails-V, the first open-source dataset with dual preference annotations for helpfulness and safety, along with multi-level safety labels (minor, moderate, severe). Additionally, we design a Multi-level Guardrail System to proactively defend against unsafe queries and adversarial attacks. By applying the Beaver-Guard-V moderation for 5 rounds of filtering and re-generation on the precursor model, the overall safety of the upstream model is significantly improved by an average of 40.9%. Experimental results demonstrate that fine-tuning different MLLMs with Safe RLHF can effectively enhance model helpfulness while ensuring improved safety. Specifically, Safe RLHF-V improves model safety by 34.2% and helpfulness by 34.3%. All of datasets, models, and code can be found at https://github.com/SafeRLHF-V to support the safety development of MLLMs and reduce potential societal risks.

With the development of large language models (LLMs), striking a balance between the performance and safety of AI systems has never been more critical. However, the inherent tension between the objectives of helpfulness and harmlessness presents a significant challenge during LLM training. To address this issue, we propose Safe Reinforcement Learning from Human Feedback (Safe RLHF), a novel algorithm for human value alignment. Safe RLHF explicitly decouples human preferences regarding helpfulness and harmlessness, effectively avoiding the crowdworkers' confusion about the tension and allowing us to train separate reward and cost models. We formalize the safety concern of LLMs as an optimization task of maximizing the reward function while satisfying specified cost constraints. Leveraging the Lagrangian method to solve this constrained problem, Safe RLHF dynamically adjusts the balance between the two objectives during fine-tuning. Through a three-round fine-tuning using Safe RLHF, we demonstrate a superior ability to mitigate harmful responses while enhancing model performance compared to existing value-aligned algorithms. Experimentally, we fine-tuned the Alpaca-7B using Safe RLHF and aligned it with collected human preferences, significantly improving its helpfulness and harmlessness according to human evaluations.

Reinforcement Learning from Human Feedback (RLHF) is a pivotal technique that aligns language models closely with human-centric values. The initial phase of RLHF involves learning human values using a reward model from ranking data. It is observed that the performance of the reward model degrades after one epoch of training, and optimizing too much against the learned reward model eventually hinders the true objective. This paper delves into these issues, leveraging the theoretical insights to design improved reward learning algorithm termed 'Iterative Data Smoothing' (IDS). The core idea is that during each training epoch, we not only update the model with the data, but also update the date using the model, replacing hard labels with soft labels. Our empirical findings highlight the superior performance of this approach over the traditional methods.

Large Language Models (LLMs) fine-tuned via Reinforcement Learning from Human Feedback (RLHF) and Reinforcement Learning with Verifiable Rewards (RLVR) significantly improve the alignment of human-AI values and further raise the upper bound of AI capabilities, particularly in reasoning-intensive, long-context Chain-of-Thought (long-CoT) tasks. However, existing RLHF (or RLVR) frameworks commonly face challenges such as inference bottlenecks and complexity barriers, restricting their accessibility for newcomers. To bridge this gap, we introduce OpenRLHF, a user-friendly, scalable, and easy-to-learn open-source RLHF framework built upon Ray, vLLM, DeepSpeed, and HuggingFace Transformers, featuring a simplified design, clear code structure, and comprehensive documentation to facilitate entry for researchers and practitioners. Experimental results show that OpenRLHF achieves superior training efficiency with speedups ranging from 1.22x to 1.68x across different model sizes compared to state-of-the-art frameworks, while requiring significantly fewer lines of code for implementation. OpenRLHF is publicly available at https://github.com/OpenRLHF/OpenRLHF, and has already been adopted by leading institutions to accelerate RLHF research and learning.

Reinforcement learning from human feedback (RLHF) is a variant of reinforcement learning (RL) that learns from human feedback instead of relying on an engineered reward function. Building on prior work on the related setting of preference-based reinforcement learning (PbRL), it stands at the intersection of artificial intelligence and human-computer interaction. This positioning offers a promising avenue to enhance the performance and adaptability of intelligent systems while also improving the alignment of their objectives with human values. The training of Large Language Models (LLMs) has impressively demonstrated this potential in recent years, where RLHF played a decisive role in targeting the model's capabilities toward human objectives. This article provides a comprehensive overview of the fundamentals of RLHF, exploring the intricate dynamics between machine agents and human input. While recent focus has been on RLHF for LLMs, our survey adopts a broader perspective, examining the diverse applications and wide-ranging impact of the technique. We delve into the core principles that underpin RLHF, shedding light on the symbiotic relationship between algorithms and human feedback, and discuss the main research trends in the field. By synthesizing the current landscape of RLHF research, this article aims to provide researchers as well as practitioners with a comprehensive understanding of this rapidly growing field of research.

Recent advances in Large Language Models (LLMs) have achieved remarkable breakthroughs in understanding and responding to user intents. However, their performance lag behind general use cases in some expertise domains, such as Chinese medicine. Existing efforts to incorporate Chinese medicine into LLMs rely on Supervised Fine-Tuning (SFT) with single-turn and distilled dialogue data. These models lack the ability for doctor-like proactive inquiry and multi-turn comprehension and cannot align responses with experts' intentions. In this work, we introduce Zhongjing, the first Chinese medical LLaMA-based LLM that implements an entire training pipeline from continuous pre-training, SFT, to Reinforcement Learning from Human Feedback (RLHF). Additionally, we construct a Chinese multi-turn medical dialogue dataset of 70,000 authentic doctor-patient dialogues, CMtMedQA, which significantly enhances the model's capability for complex dialogue and proactive inquiry initiation. We also define a refined annotation rule and evaluation criteria given the unique characteristics of the biomedical domain. Extensive experimental results show that Zhongjing outperforms baselines in various capacities and matches the performance of ChatGPT in some abilities, despite the 100x parameters. Ablation studies also demonstrate the contributions of each component: pre-training enhances medical knowledge, and RLHF further improves instruction-following ability and safety. Our code, datasets, and models are available at https://github.com/SupritYoung/Zhongjing.

Reinforcement Learning from Human Feedback (RLHF) is a widely adopted approach for aligning large language models with human values. However, RLHF relies on a reward model that is trained with a limited amount of human preference data, which could lead to inaccurate predictions. As a result, RLHF may produce outputs that are misaligned with human values. To mitigate this issue, we contribute a reward ensemble method that allows the reward model to make more accurate predictions. As using an ensemble of large language model-based reward models can be computationally and resource-expensive, we explore efficient ensemble methods including linear-layer ensemble and LoRA-based ensemble. Empirically, we run Best-of-n and Proximal Policy Optimization with our ensembled reward models, and verify that our ensemble methods help improve the alignment performance of RLHF outputs.

This study explores the scaling properties of Reinforcement Learning from Human Feedback (RLHF) in Large Language Models (LLMs). Although RLHF is considered an important step in post-training of LLMs, its scaling potential is still largely unknown. We systematically analyze key components in the RLHF framework--model size, data composition, and inference budget--and their impacts on performance. Our findings show that increasing data diversity and volume improves reward model performance, helping process-supervision models scale better. For policy training, more response samples per prompt boost performance initially but quickly plateau. And larger reward models offer modest gains in policy training. In addition, larger policy models benefit less from RLHF with a fixed reward model. Overall, RLHF scales less efficiently than pretraining, with diminishing returns from additional computational resources. Based on these observations, we propose strategies to optimize RLHF performance within computational limits.

Reinforcement Learning from Human Feedback (RLHF) is a popular method for aligning Language Models (LM) with human values and preferences. RLHF requires a large number of preference pairs as training data, which are often used in both the Supervised Fine-Tuning and Reward Model training and therefore publicly available datasets are commonly used. In this work, we study to what extent a malicious actor can manipulate the LMs generations by poisoning the preferences, i.e., injecting poisonous preference pairs into these datasets and the RLHF training process. We propose strategies to build poisonous preference pairs and test their performance by poisoning two widely used preference datasets. Our results show that preference poisoning is highly effective: injecting a small amount of poisonous data (1-5\% of the original dataset), we can effectively manipulate the LM to generate a target entity in a target sentiment (positive or negative). The findings from our experiments also shed light on strategies to defend against the preference poisoning attack.

Reinforcement Learning from Human Feedback (RLHF) facilitates the alignment of large language models with human preferences, significantly enhancing the quality of interactions between humans and these models. InstructGPT implements RLHF through several stages, including Supervised Fine-Tuning (SFT), reward model training, and Proximal Policy Optimization (PPO). PPO, however, is sensitive to hyperparameters and requires a minimum of four models in its standard implementation, which makes it hard to train. In contrast, we propose a novel learning paradigm called RRHF, which scores responses generated by different sampling policies and learns to align them with human preferences through ranking loss. RRHF can efficiently align language model output probabilities with human preferences as robust as fine-tuning and it only needs 1 to 2 models during tuning. In addition, RRHF can be considered an extension of SFT and reward models while being simpler than PPO in terms of coding, model counts, and hyperparameters. The entire alignment process can be accomplished within a single RRHF training session. We evaluate RRHF using LLaMA and Alpaca on Helpful and Harmless data, demonstrating performance comparable to PPO.

Reinforcement Learning from Human Feedback (RLHF) is a crucial technique for aligning language models with human preferences, playing a pivotal role in the success of conversational models like GPT-4, ChatGPT, and Llama 2. A core challenge in employing RLHF lies in training a reliable reward model (RM), which relies on high-quality labels typically provided by human experts or advanced AI system. These methods can be costly and may introduce biases that affect the language model's responses. As language models improve, human input may become less effective in further enhancing their performance. In this paper, we propose Self-Evolved Reward Learning (SER), a novel approach where the RM generates additional training data to iteratively improve itself. We conducted extensive experiments on multiple datasets such as HH-RLHF and UltraFeedback, using models like Mistral and Llama 3, and compare SER against various baselines. Our results demonstrate that even with limited human-annotated data, learning from self-feedback can robustly enhance RM performance, thereby boosting the capabilities of large language models (LLMs).

Large Language Models (LLMs) have revolutionized natural language processing, yet aligning these models with human values and preferences using RLHF remains a significant challenge. This challenge is characterized by various instabilities, such as reward hacking and catastrophic forgetting. In this technical report, we propose two innovations to stabilize RLHF training: 1) Advantage Model, which directly models advantage score i.e., extra reward compared to the expected rewards and regulates score distributions across tasks to prevent reward hacking. 2) Selective Rehearsal, which mitigates catastrophic forgetting by strategically selecting data for PPO training and knowledge rehearsing. Our experimental analysis on public and proprietary datasets reveals that the proposed methods not only increase stability in RLHF training but also achieve higher reward scores and win rates.

Reinforcement Learning from Human Feedback (RLHF) has proven to be a strong method to align Pretrained Large Language Models (LLMs) with human preferences. But training models with RLHF is computationally expensive, and an overall complex process. In this work, we study RLHF where the underlying models are trained using the parameter efficient method of Low-Rank Adaptation (LoRA) introduced by Hu et al. [2021]. We investigate the setup of "Parameter Efficient Reinforcement Learning" (PERL), in which we perform reward model training and reinforcement learning using LoRA. We compare PERL to conventional fine-tuning (full-tuning) across various configurations for 7 benchmarks, including 2 novel datasets, of reward modeling and reinforcement learning. We find that PERL performs on par with the conventional RLHF setting, while training faster, and with less memory. This enables the high performance of RLHF, while reducing the computational burden that limits its adoption as an alignment technique for Large Language Models. We also release 2 novel thumbs up/down preference datasets: "Taskmaster Coffee", and "Taskmaster Ticketing" to promote research around RLHF.

Reinforcement learning from human feedback (RLHF) is one of the key techniques that helps large language models (LLMs) to follow instructions and provide helpful and harmless responses. While direct policy optimization methods exist, state-of-the-art LLMs adopt RL-based methods (usually PPO) in RLHF to train the policy to generate good responses guided by a reward model learned from preference data. The main challenge of these methods is the inaccuracy of the intermediate reward model, especially in code generation tasks that require long and complex reasoning to score a response. We find that the reliability of the reward model varies across responses assigned with different rewards. This motivates us to filter the samples whose rewards may be unreliable to improve signal-to-noise ratio during policy learning, resulting in Policy Filtration for Proximal Policy Optimization (PF-PPO). To choose a proper policy filtration strategy for a given reward model, the coefficient of determination (R^2) between rewards and actual scores on filtered samples serves as a good metrics and helps us find several promising strategies. We provide extensive experiments to validate the effectiveness of PF-PPO in code generation tasks, and find that some variants of PF-PPO are highly effective and achieve new state-of-the-art performance across 7-billion-parameter models on HumanEval, MBPP, and a new and more challenging LeetCode Contest benchmark.

Reinforcement learning from human feedback (RLHF) has emerged as a powerful technique to make large language models (LLMs) more capable in complex settings. RLHF proceeds as collecting human preference data, training a reward model on said data, and optimizing a base ML model with respect to said reward for extrinsic evaluation metrics (e.g. MMLU, GSM8k). RLHF relies on many assumptions about how the various pieces fit together, such as a reward model capturing human preferences and an RL optimizer extracting the right signal from a reward model. As the RLHF process involves many distinct design decisions, it is easy to assume that multiple processes are correlated and therefore numerically linked. This apparent correlation is often not true, where reward models are easily overoptimized or RL optimizers can reduce performance on tasks not modeled in the data. Notable manifestations of models trained with imperfect RLHF systems are those that are prone to refusing basic requests for safety reasons or appearing lazy in generations. As chat model evaluation becomes increasingly nuanced, the reliance on a perceived link between reward model training, RL scores, and downstream performance drives these issues, which we describe as an objective mismatch. In this paper, we illustrate the causes of this issue, reviewing relevant literature from model-based reinforcement learning, and argue for solutions. By solving objective mismatch in RLHF, the ML models of the future will be more precisely aligned to user instructions for both safety and helpfulness.

Objective: The reading level of health educational materials significantly influences the understandability and accessibility of the information, particularly for minoritized populations. Many patient educational resources surpass the reading level and complexity of widely accepted standards. There is a critical need for high-performing text simplification models in health information to enhance dissemination and literacy. This need is particularly acute in cancer education, where effective prevention and screening education can substantially reduce morbidity and mortality. Methods: We introduce Simplified Digestive Cancer (SimpleDC), a parallel corpus of cancer education materials tailored for health text simplification research, comprising educational content from the American Cancer Society, Centers for Disease Control and Prevention, and National Cancer Institute. Utilizing SimpleDC alongside the existing Med-EASi corpus, we explore Large Language Model (LLM)-based simplification methods, including fine-tuning, reinforcement learning (RL), reinforcement learning with human feedback (RLHF), domain adaptation, and prompt-based approaches. Our experimentation encompasses Llama 2 and GPT-4. A novel RLHF reward function is introduced, featuring a lightweight model adept at distinguishing between original and simplified texts, thereby enhancing the model's effectiveness with unlabeled data. Results: Fine-tuned Llama 2 models demonstrated high performance across various metrics. Our innovative RLHF reward function surpassed existing RL text simplification reward functions in effectiveness. The results underscore that RL/RLHF can augment fine-tuning, facilitating model training on unlabeled text and improving performance.

Reinforcement learning from human feedback (RLHF) has become an important technical and storytelling tool to deploy the latest machine learning systems. In this book, we hope to give a gentle introduction to the core methods for people with some level of quantitative background. The book starts with the origins of RLHF -- both in recent literature and in a convergence of disparate fields of science in economics, philosophy, and optimal control. We then set the stage with definitions, problem formulation, data collection, and other common math used in the literature. The core of the book details every optimization stage in using RLHF, from starting with instruction tuning to training a reward model and finally all of rejection sampling, reinforcement learning, and direct alignment algorithms. The book concludes with advanced topics -- understudied research questions in synthetic data and evaluation -- and open questions for the field.

Recent advancements in Large Language Models (LLMs) have garnered wide attention and led to successful products such as ChatGPT and GPT-4. Their proficiency in adhering to instructions and delivering harmless, helpful, and honest (3H) responses can largely be attributed to the technique of Reinforcement Learning from Human Feedback (RLHF). In this paper, we aim to link the research in conventional RL to RL techniques used in LLM research. Demystify this technique by discussing why, when, and how RL excels. Furthermore, we explore potential future avenues that could either benefit from or contribute to RLHF research. Highlighted Takeaways: 1. RLHF is Online Inverse RL with Offline Demonstration Data. 2. RLHF > SFT because Imitation Learning (and Inverse RL) > Behavior Cloning (BC) by alleviating the problem of compounding error. 3. The RM step in RLHF generates a proxy of the expensive human feedback, such an insight can be generalized to other LLM tasks such as prompting evaluation and optimization where feedback is also expensive. 4. The policy learning in RLHF is more challenging than conventional problems studied in IRL due to their high action dimensionality and feedback sparsity. 5. The main superiority of PPO over off-policy value-based methods is its stability gained from (almost) on-policy data and conservative policy updates.

ChatGPT-like models have revolutionized various applications in artificial intelligence, from summarization and coding to translation, matching or even surpassing human performance. However, the current landscape lacks an accessible, efficient, and cost-effective end-to-end RLHF (Reinforcement Learning with Human Feedback) training pipeline for these powerful models, particularly when training at the scale of billions of parameters. This paper introduces DeepSpeed-Chat, a novel system that democratizes RLHF training, making it accessible to the AI community. DeepSpeed-Chat offers three key capabilities: an easy-to-use training and inference experience for ChatGPT-like models, a DeepSpeed-RLHF pipeline that replicates the training pipeline from InstructGPT, and a robust DeepSpeed-RLHF system that combines various optimizations for training and inference in a unified way. The system delivers unparalleled efficiency and scalability, enabling training of models with hundreds of billions of parameters in record time and at a fraction of the cost. With this development, DeepSpeed-Chat paves the way for broader access to advanced RLHF training, even for data scientists with limited resources, thereby fostering innovation and further development in the field of AI.

Large language models (LLMs) have formulated a blueprint for the advancement of artificial general intelligence. Its primary objective is to function as a human-centric (helpful, honest, and harmless) assistant. Alignment with humans assumes paramount significance, and reinforcement learning with human feedback (RLHF) emerges as the pivotal technological paradigm underpinning this pursuit. Current technical routes usually include reward models to measure human preferences, Proximal Policy Optimization (PPO) to optimize policy model outputs, and process supervision to improve step-by-step reasoning capabilities. However, due to the challenges of reward design, environment interaction, and agent training, coupled with huge trial and error cost of large language models, there is a significant barrier for AI researchers to motivate the development of technical alignment and safe landing of LLMs. The stable training of RLHF has still been a puzzle. In the first report, we dissect the framework of RLHF, re-evaluate the inner workings of PPO, and explore how the parts comprising PPO algorithms impact policy agent training. We identify policy constraints being the key factor for the effective implementation of the PPO algorithm. Therefore, we explore the PPO-max, an advanced version of PPO algorithm, to efficiently improve the training stability of the policy model. Based on our main results, we perform a comprehensive analysis of RLHF abilities compared with SFT models and ChatGPT. The absence of open-source implementations has posed significant challenges to the investigation of LLMs alignment. Therefore, we are eager to release technical reports, reward models and PPO codes

Abstract Language Models (LMs) achieve substantial language capabilities when finetuned using Reinforcement Learning with Human Feedback (RLHF). However, RLHF is an unstable and data-hungry process that continually requires new high-quality LM-generated data for finetuning. We introduce Advantage-Leftover Lunch RL (A-LoL), a new class of offline policy gradient algorithms that enable RL training on any pre-existing data. By assuming the entire LM output sequence as a single action, A-LoL allows incorporating sequence-level classifiers or human-designed scoring functions as rewards. Subsequently, by using LM's internal sequence-level value estimate, A-LoL filters negative advantage (low-quality) data points during training, making it resilient to noise. Overall, A-LoL is an easy-to-implement LM training recipe that is sample-efficient and stable. We demonstrate the effectiveness of A-LoL and its variants with a set of four different language generation tasks. We compare against both online RL (PPO) and recent preference-based (DPO, PRO) and reward-based (GOLD) offline RL baselines. On the commonly-used RLHF benchmark, Helpful and Harmless Assistant (HHA), LMs trained with A-LoL methods achieve the highest diversity while also being rated more safe and helpful than baselines according to humans. Additionally, in the remaining three tasks, A-LoL could optimize multiple distinct reward functions even when using noisy or suboptimal training data. We also release our experimental code. https://github.com/abaheti95/LoL-RL

Optimization (PPO) has been positioned by recent literature as the canonical method for the RL part of RLHF. PPO performs well empirically but has a heuristic motivation and handles the KL-divergence constraint used in LM-RLHF in an ad-hoc manner and suffers form reward oscillations, entropy collapse, value function drift, and sudden policy divergence that require frequent restarts and extensive hyperparameter tuning. In this paper, we develop a new pure on policy actor-critic RL method for the LM-RLHF setting. We present SAFE (Stable Alignment Finetuning with Entropy-aware control),a novel RLHF algorithm that combines a Double Soft-Min Critic for pessimistic value estimation with a new multi-layer stabilization framework combining entropy-gated KL regulation, and PID-controlled adaptive thresholds. Unlike standard PPO's symmetric KL penalties, SAFE distinguishes high-entropy exploration from low-entropy mode collapse and adjusts penalties dynamically based on reward velocity. Experiments on a 3B parameter model show SAFE achieves +5.15\% training-average reward than PPO (0.725 vs 0.689), negligible reward crashes, and superior KL control than ppo . Our method adds minimal computational overhead and provides an interpretable, crash-resistant RLHF framework that maintains aggressive learning speed while ensuring stable long-horizon optimization suitable for production deployment. Code is available at https://github.com/ryyzn9/SAFE

Reinforcement learning from human feedback (RLHF) is a standard approach for fine-tuning large language models to follow instructions. As part of this process, learned reward models are used to approximately model human preferences. However, as imperfect representations of the "true" reward, these learned reward models are susceptible to overoptimization. Gao et al. (2023) studied this phenomenon in a synthetic human feedback setup with a significantly larger "gold" reward model acting as the true reward (instead of humans) and showed that overoptimization remains a persistent problem regardless of the size of the proxy reward model and training data used. Using a similar setup, we conduct a systematic study to evaluate the efficacy of using ensemble-based conservative optimization objectives, specifically worst-case optimization (WCO) and uncertainty-weighted optimization (UWO), for mitigating reward model overoptimization when using two optimization methods: (a) best-of-n sampling (BoN) (b) proximal policy optimization (PPO). We additionally extend the setup of Gao et al. (2023) to include 25% label noise to better mirror real-world conditions. Both with and without label noise, we find that conservative optimization practically eliminates overoptimization and improves performance by up to 70% for BoN sampling. For PPO, ensemble-based conservative optimization always reduces overoptimization and outperforms single reward model optimization. Moreover, combining it with a small KL penalty successfully prevents overoptimization at no performance cost. Overall, our results demonstrate that ensemble-based conservative optimization can effectively counter overoptimization.

Reinforcement learning from human feedback (RLHF) has emerged as a powerful technique to make large language models (LLMs) easier to use and more effective. A core piece of the RLHF process is the training and utilization of a model of human preferences that acts as a reward function for optimization. This approach, which operates at the intersection of many stakeholders and academic disciplines, remains poorly understood. RLHF reward models are often cited as being central to achieving performance, yet very few descriptors of capabilities, evaluations, training methods, or open-source models exist. Given this lack of information, further study and transparency is needed for learned RLHF reward models. In this paper, we illustrate the complex history of optimizing preferences, and articulate lines of inquiry to understand the sociotechnical context of reward models. In particular, we highlight the ontological differences between costs, rewards, and preferences at stake in RLHF's foundations, related methodological tensions, and possible research directions to improve general understanding of how reward models function.

Reinforcement learning from human feedback (RLHF) has emerged as an effective approach to aligning large language models (LLMs) to human preferences. RLHF contains three steps, i.e., human preference collecting, reward learning, and policy optimization, which are usually performed serially. Despite its popularity, however, (fixed) reward models may suffer from inaccurate off-distribution, since policy optimization continuously shifts LLMs' data distribution. Repeatedly collecting new preference data from the latest LLMs may alleviate this issue, which unfortunately makes the resulting system more complicated and difficult to optimize. In this paper, we propose reward learning on policy (RLP), an unsupervised framework that refines a reward model using policy samples to keep it on-distribution. Specifically, an unsupervised multi-view learning method is introduced to learn robust representations of policy samples. Meanwhile, a synthetic preference generation approach is developed to simulate high-quality preference data with policy outputs. Extensive experiments on three benchmark datasets show that RLP consistently outperforms the state-of-the-art. Our code is available at https://github.com/AlibabaResearch/DAMO-ConvAI/tree/main/rlp.

The development of Large Language Models (LLMs) often confronts challenges stemming from the heavy reliance on human annotators in the reinforcement learning with human feedback (RLHF) framework, or the frequent and costly external queries tied to the self-instruct paradigm. In this work, we pivot to Reinforcement Learning (RL) -- but with a twist. Diverging from the typical RLHF, which refines LLMs following instruction data training, we use RL to directly generate the foundational instruction dataset that alone suffices for fine-tuning. Our method, TeaMs-RL, uses a suite of textual operations and rules, prioritizing the diversification of training datasets. It facilitates the generation of high-quality data without excessive reliance on external advanced models, paving the way for a single fine-tuning step and negating the need for subsequent RLHF stages. Our findings highlight key advantages of our approach: reduced need for human involvement and fewer model queries (only 5.73% of the strong baseline's total), along with enhanced capabilities of LLMs in crafting and comprehending complex instructions compared to strong baselines, and substantially improved model privacy protection. Code is available at the link: https://github.com/SafeRL-Lab/TeaMs-RL

Reinforcement learning from human feedback (RLHF) is a crucial technique in aligning large language models (LLMs) with human preferences, ensuring these LLMs behave in beneficial and comprehensible ways to users. However, a longstanding challenge in human alignment techniques based on reinforcement learning lies in their inherent complexity and difficulty in training. To address this challenge, we present a simple yet effective Contrastive Learning Framework for Human Alignment (CLHA) to align LLMs with human preferences directly. CLHA employs a novel rescoring strategy to evaluate the noise within the data by considering its inherent quality and dynamically adjusting the training process. Simultaneously, CLHA utilizes pairwise contrastive loss and adaptive supervised fine-tuning loss to adaptively modify the likelihood of generating responses, ensuring enhanced alignment with human preferences. Using advanced methods, CLHA surpasses other algorithms, showcasing superior performance in terms of reward model scores, automatic evaluations, and human assessments on the widely used ``Helpful and Harmless'' dataset.

Reinforcement learning from human feedback (RLHF) is a popular strategy for aligning large language models (LLMs) with desired behaviors. Reward modeling is a crucial step in RLHF. However, collecting paired preference data for training reward models is often costly and time-consuming, especially for domain-specific preferences requiring expert annotation. To address this challenge, we propose the Domain knowledge merged Reward Model (DogeRM), a novel framework that integrates domain-specific knowledge into a general reward model by model merging. The experiments demonstrate that DogeRM enhances performance across different benchmarks and provide a detailed analysis showcasing the effects of model merging, showing the great potential of facilitating model alignment.

Reinforcement learning from human feedback (RLHF) has become the leading approach for fine-tuning large language models (LLM). However, RLHF has limitations in multi-task learning (MTL) due to challenges of reward hacking and extreme multi-objective optimization (i.e., trade-off of multiple and/or sometimes conflicting objectives). Applying RLHF for MTL currently requires careful tuning of the weights for reward model and data combinations. This is often done via human intuition and does not generalize. In this work, we introduce a novel post-training paradigm which we called Constrained Generative Policy Optimization (CGPO). The core of CGPO is Mixture of Judges (MoJ) with cost-efficient constrained policy optimization with stratification, which can identify the perfect blend in RLHF in a principled manner. It shows strong empirical results with theoretical guarantees, does not require extensive hyper-parameter tuning, and is plug-and-play in common post-training pipelines. Together, this can detect and mitigate reward hacking behaviors while reaching a pareto-optimal point across an extremely large number of objectives. Our empirical evaluations demonstrate that CGPO significantly outperforms standard RLHF algorithms like PPO and DPO across various tasks including general chat, STEM questions, instruction following, and coding. Specifically, CGPO shows improvements of 7.4% in AlpacaEval-2 (general chat), 12.5% in Arena-Hard (STEM & reasoning), and consistent gains in other domains like math and coding. Notably, PPO, while commonly used, is prone to severe reward hacking in popular coding benchmarks, which CGPO successfully addresses. This breakthrough in RLHF not only tackles reward hacking and extreme multi-objective optimization challenges but also advances the state-of-the-art in aligning general-purpose LLMs for diverse applications.

As large language models (LLMs) have increased in their capabilities, so does their potential for dual use. To reduce harmful outputs, produces and vendors of LLMs have used reinforcement learning with human feedback (RLHF). In tandem, LLM vendors have been increasingly enabling fine-tuning of their most powerful models. However, concurrent work has shown that fine-tuning can remove RLHF protections. We may expect that the most powerful models currently available (GPT-4) are less susceptible to fine-tuning attacks. In this work, we show the contrary: fine-tuning allows attackers to remove RLHF protections with as few as 340 examples and a 95% success rate. These training examples can be automatically generated with weaker models. We further show that removing RLHF protections does not decrease usefulness on non-censored outputs, providing evidence that our fine-tuning strategy does not decrease usefulness despite using weaker models to generate training data. Our results show the need for further research on protections on LLMs.

Reinforcement Learning from Human Feedback (RLHF) is widely used in Large Language Model (LLM) alignment. Traditional RL can be modeled as a dataflow, where each node represents computation of a neural network (NN) and each edge denotes data dependencies between the NNs. RLHF complicates the dataflow by expanding each node into a distributed LLM training or generation program, and each edge into a many-to-many multicast. Traditional RL frameworks execute the dataflow using a single controller to instruct both intra-node computation and inter-node communication, which can be inefficient in RLHF due to large control dispatch overhead for distributed intra-node computation. Existing RLHF systems adopt a multi-controller paradigm, which can be inflexible due to nesting distributed computation and data communication. We propose HybridFlow, which combines single-controller and multi-controller paradigms in a hybrid manner to enable flexible representation and efficient execution of the RLHF dataflow. We carefully design a set of hierarchical APIs that decouple and encapsulate computation and data dependencies in the complex RLHF dataflow, allowing efficient operation orchestration to implement RLHF algorithms and flexible mapping of the computation onto various devices. We further design a 3D-HybridEngine for efficient actor model resharding between training and generation phases, with zero memory redundancy and significantly reduced communication overhead. Our experimental results demonstrate 1.53times~20.57times throughput improvement when running various RLHF algorithms using HybridFlow, as compared with state-of-the-art baselines. HybridFlow source code will be available at https://github.com/volcengine/verl.

Language models (LMs) can produce errors that are hard to detect for humans, especially when the task is complex. RLHF, the most popular post-training method, may exacerbate this problem: to achieve higher rewards, LMs might get better at convincing humans that they are right even when they are wrong. We study this phenomenon under a standard RLHF pipeline, calling it "U-SOPHISTRY" since it is Unintended by model developers. Specifically, we ask time-constrained (e.g., 3-10 minutes) human subjects to evaluate the correctness of model outputs and calculate humans' accuracy against gold labels. On a question-answering task (QuALITY) and programming task (APPS), RLHF makes LMs better at convincing our subjects but not at completing the task correctly. RLHF also makes the model harder to evaluate: our subjects' false positive rate increases by 24.1% on QuALITY and 18.3% on APPS. Finally, we show that probing, a state-of-the-art approach for detecting Intended Sophistry (e.g. backdoored LMs), does not generalize to U-SOPHISTRY. Our results highlight an important failure mode of RLHF and call for more research in assisting humans to align them.

Reinforcement Learning from Human Feedback (RLHF) is an effective approach for aligning language models to human preferences. Central to RLHF is learning a reward function for scoring human preferences. Two main approaches for learning a reward model are 1) training an EXplicit Reward Model (EXRM) as in RLHF, and 2) using an implicit reward learned from preference data through methods such as Direct Preference Optimization (DPO). Prior work has shown that the implicit reward model of DPO (denoted as DPORM) can approximate an EXRM in the limit. DPORM's effectiveness directly implies the optimality of the learned policy, and also has practical implication for LLM alignment methods including iterative DPO. However, it is unclear how well DPORM empirically matches the performance of EXRM. This work studies the accuracy at distinguishing preferred and rejected answers for both DPORM and EXRM. Our findings indicate that even though DPORM fits the training dataset comparably, it generalizes less effectively than EXRM, especially when the validation datasets contain distribution shifts. Across five out-of-distribution settings, DPORM has a mean drop in accuracy of 3% and a maximum drop of 7%. These findings highlight that DPORM has limited generalization ability and substantiates the integration of an explicit reward model in iterative DPO approaches.

Reinforcement Learning with Human Feedback (RLHF) has been demonstrated to significantly enhance the performance of large language models (LLMs) by aligning their outputs with desired human values through instruction tuning. However, RLHF is constrained by the expertise and productivity limitations of human evaluators. A response to this downside is to fall back to supervised fine-tuning (SFT) with additional carefully selected expert demonstrations. However, while this method has been proven to be effective, it invariably also leads to increased human-in-the-loop overhead. In this study, we propose another alternative approach: Reinforcement Learning with Generative Adversarial Feedback (RLGAF) to RLHF and SFT, which uses a generative adversarial training style to enable the LLMs to learn useful human expert demonstrations without being directly exposed to the training examples, thus enabling good generalization capabilities while preserving sample efficiency. Our preliminary findings indicate that RLGAF can help align LLMs outputs with competitive performance against RLHF and SFT, while not suffering from their respective inherent restrictions, suggesting promising avenues for further research on automating AI alignment.

We introduce a new benchmark for reward models that quantifies their ability to produce strong language models through RLHF (Reinforcement Learning from Human Feedback). The gold-standard approach is to run a full RLHF training pipeline and directly probe downstream LLM performance. However, this process is prohibitively expensive. To address this, we build a predictive model of downstream LLM performance by evaluating the reward model on proxy tasks. These proxy tasks consist of a large-scale human preference and a verifiable correctness preference dataset, in which we measure 12 metrics across 12 domains. To investigate which reward model metrics are most correlated to gold-standard RLHF outcomes, we launch an end-to-end RLHF experiment on a large-scale crowdsourced human preference platform to view real reward model downstream performance as ground truth. Ultimately, we compile our data and findings into Preference Proxy Evaluations (PPE), the first reward model benchmark explicitly linked to post-RLHF real-world human preference performance, which we open-source for public use and further development. Our code and evaluations can be found at https://github.com/lmarena/PPE .

We introduce SafeWork-R1, a cutting-edge multimodal reasoning model that demonstrates the coevolution of capabilities and safety. It is developed by our proposed SafeLadder framework, which incorporates large-scale, progressive, safety-oriented reinforcement learning post-training, supported by a suite of multi-principled verifiers. Unlike previous alignment methods such as RLHF that simply learn human preferences, SafeLadder enables SafeWork-R1 to develop intrinsic safety reasoning and self-reflection abilities, giving rise to safety `aha' moments. Notably, SafeWork-R1 achieves an average improvement of 46.54% over its base model Qwen2.5-VL-72B on safety-related benchmarks without compromising general capabilities, and delivers state-of-the-art safety performance compared to leading proprietary models such as GPT-4.1 and Claude Opus 4. To further bolster its reliability, we implement two distinct inference-time intervention methods and a deliberative search mechanism, enforcing step-level verification. Finally, we further develop SafeWork-R1-InternVL3-78B, SafeWork-R1-DeepSeek-70B, and SafeWork-R1-Qwen2.5VL-7B. All resulting models demonstrate that safety and capability can co-evolve synergistically, highlighting the generalizability of our framework in building robust, reliable, and trustworthy general-purpose AI.

Reinforcement learning from human feedback (RLHF) has demonstrated effectiveness in aligning large language models (LLMs) with human preferences. However, token-level RLHF suffers from the credit assignment problem over long sequences, where delayed rewards make it challenging for the model to discern which actions contributed to successful outcomes. This hinders learning efficiency and slows convergence. In this paper, we propose MA-RLHF, a simple yet effective RLHF framework that incorporates macro actions -- sequences of tokens or higher-level language constructs -- into the learning process. By operating at this higher level of abstraction, our approach reduces the temporal distance between actions and rewards, facilitating faster and more accurate credit assignment. This results in more stable policy gradient estimates and enhances learning efficiency within each episode, all without increasing computational complexity during training or inference. We validate our approach through extensive experiments across various model sizes and tasks, including text summarization, dialogue generation, question answering, and program synthesis. Our method achieves substantial performance improvements over standard RLHF, with performance gains of up to 30% in text summarization and code generation, 18% in dialogue, and 8% in question answering tasks. Notably, our approach reaches parity with vanilla RLHF 1.7x to 2x faster in terms of training time and continues to outperform it with further training. We will make our code and data publicly available at https://github.com/ernie-research/MA-RLHF .

Reinforcement Learning from Human Feedback (RLHF) stands as a pivotal technique in empowering large language model (LLM) applications. Since RLHF involves diverse computational workloads and intricate dependencies among multiple LLMs, directly adopting parallelization techniques from supervised training can result in sub-optimal performance. To overcome this limitation, we propose a novel approach named parameter ReaLlocation, which dynamically redistributes LLM parameters in the cluster and adapts parallelization strategies during training. Building upon this idea, we introduce ReaLHF, a pioneering system capable of automatically discovering and running efficient execution plans for RLHF training given the desired algorithmic and hardware configurations. ReaLHF formulates the execution plan for RLHF as an augmented dataflow graph. Based on this formulation, ReaLHF employs a tailored search algorithm with a lightweight cost estimator to discover an efficient execution plan. Subsequently, the runtime engine deploys the selected plan by effectively parallelizing computations and redistributing parameters. We evaluate ReaLHF on the LLaMA-2 models with up to 4times70 billion parameters and 128 GPUs. The experiment results showcase ReaLHF's substantial speedups of 2.0-10.6times compared to baselines. Furthermore, the execution plans generated by ReaLHF exhibit an average of 26% performance improvement over heuristic approaches based on Megatron-LM. The source code of ReaLHF is publicly available at https://github.com/openpsi-project/ReaLHF .

Reinforcement learning from human feedback (RLHF) is fundamentally limited by the capacity of humans to correctly evaluate model output. To improve human evaluation ability and overcome that limitation this work trains "critic" models that help humans to more accurately evaluate model-written code. These critics are themselves LLMs trained with RLHF to write natural language feedback highlighting problems in code from real-world assistant tasks. On code containing naturally occurring LLM errors model-written critiques are preferred over human critiques in 63% of cases, and human evaluation finds that models catch more bugs than human contractors paid for code review. We further confirm that our fine-tuned LLM critics can successfully identify hundreds of errors in ChatGPT training data rated as "flawless", even though the majority of those tasks are non-code tasks and thus out-of-distribution for the critic model. Critics can have limitations of their own, including hallucinated bugs that could mislead humans into making mistakes they might have otherwise avoided, but human-machine teams of critics and contractors catch similar numbers of bugs to LLM critics while hallucinating less than LLMs alone.

We present 360Zhinao models with 7B parameter size and context lengths spanning 4K, 32K and 360K, all available at https://github.com/Qihoo360/360zhinao. For rapid development in pretraining, we establish a stable and sensitive ablation environment to evaluate and compare experiment runs with minimal model size. Under such guidance, we perfect our data cleaning and composition strategies to pretrain 360Zhinao-7B-Base on 3.4T tokens. We also mainly emphasize data during alignment, where we strive to balance quantity and quality with filtering and reformatting. With tailored data, 360Zhinao-7B's context window is easily extended to 32K and 360K. RMs and RLHF are trained following SFT and credibly applied to specific tasks. All together these contributions lead to 360Zhinao-7B's competitive performance among models of similar size.

Fine-tuning large language models (LLMs) on human preferences, typically through reinforcement learning from human feedback (RLHF), has proven successful in enhancing their capabilities. However, ensuring the safety of LLMs during the fine-tuning remains a critical concern, and mitigating the potential conflicts in safety and helpfulness is costly in RLHF. To address this issue, we propose a supervised learning framework called Bi-Factorial Preference Optimization (BFPO), which re-parameterizes a joint RLHF objective of both safety and helpfulness into a single supervised learning objective. In the supervised optimization, a labeling function is used to capture global preferences ranking to balance both safety and helpfulness. To evaluate BFPO, we develop a benchmark including comprehensive discriminative and generative tasks for helpfulness and harmlessness. The results indicate that our method significantly outperforms existing approaches in both safety and helpfulness. Moreover, BFPO eliminates the need for human prompting and annotation in LLM fine-tuning while achieving the same level of safety as methods that heavily rely on human labor, with less than 10% of the computational resources. The training recipes and models will be released.

Reinforcement Learning from Human Feedback (RLHF) has emerged as a important paradigm for aligning large language models (LLMs) with human preferences during post-training. This framework typically involves two stages: first, training a reward model on human preference data, followed by optimizing the language model using reinforcement learning algorithms. However, current RLHF approaches may constrained by two limitations. First, existing RLHF frameworks often rely on Bradley-Terry models to assign scalar rewards based on pairwise comparisons of individual responses. However, this approach imposes significant challenges on reward model (RM), as the inherent variability in prompt-response pairs across different contexts demands robust calibration capabilities from the RM. Second, reward models are typically initialized from generative foundation models, such as pre-trained or supervised fine-tuned models, despite the fact that reward models perform discriminative tasks, creating a mismatch. This paper introduces Pairwise-RL, a RLHF framework that addresses these challenges through a combination of generative reward modeling and a pairwise proximal policy optimization (PPO) algorithm. Pairwise-RL unifies reward model training and its application during reinforcement learning within a consistent pairwise paradigm, leveraging generative modeling techniques to enhance reward model performance and score calibration. Experimental evaluations demonstrate that Pairwise-RL outperforms traditional RLHF frameworks across both internal evaluation datasets and standard public benchmarks, underscoring its effectiveness in improving alignment and model behavior.

Reinforcement Learning from Human Feedback (RLHF) has emerged as a critical approach for aligning large language models with human preferences, witnessing rapid algorithmic evolution through methods such as Proximal Policy Optimization (PPO), Direct Preference Optimization (DPO), REINFORCE Leave One-Out (RLOO), ReMax, and Group Relative Policy Optimization (GRPO). We present REINFORCE++, an enhanced variant of the classical REINFORCE algorithm that incorporates key optimization techniques from PPO while eliminating the need for a critic network. REINFORCE++ achieves three primary objectives: (1) simplicity (2) enhanced training stability, and (3) reduced computational overhead. Through extensive empirical evaluation, we demonstrate that REINFORCE++ exhibits superior stability compared to GRPO and achieves greater computational efficiency than PPO while maintaining comparable performance. The implementation is available at https://github.com/OpenRLHF/OpenRLHF.

Reinforcement learning with human feedback (RLHF) has become the dominant method to align large models to user preferences. Unlike fine-tuning, for which there are many studies regarding training data memorization, it is not clear how memorization is affected by or introduced in the RLHF alignment process. Understanding this relationship is important as real user data may be collected and used to align large models; if user data is memorized during RLHF and later regurgitated, this could raise privacy concerns. In this work, we analyze how training data memorization can surface and propagate through each phase of RLHF. We focus our study on code completion models, as code completion is one of the most popular use cases for large language models. We find that RLHF significantly decreases the chance that data used for reward modeling and reinforcement learning is memorized, in comparison to aligning via directly fine-tuning on this data, but that examples already memorized during the fine-tuning stage of RLHF, will, in the majority of cases, remain memorized after RLHF.

While large language models demonstrate remarkable capabilities, they often present challenges in terms of safety, alignment with human values, and stability during training. Here, we focus on two prevalent methods used to align these models, Supervised Fine-Tuning (SFT) and Reinforcement Learning from Human Feedback (RLHF). SFT is simple and robust, powering a host of open-source models, while RLHF is a more sophisticated method used in top-tier models like ChatGPT but also suffers from instability and susceptibility to reward hacking. We propose a novel approach, Supervised Iterative Learning from Human Feedback (SuperHF), which seeks to leverage the strengths of both methods. Our hypothesis is two-fold: that the reward model used in RLHF is critical for efficient data use and model generalization and that the use of Proximal Policy Optimization (PPO) in RLHF may not be necessary and could contribute to instability issues. SuperHF replaces PPO with a simple supervised loss and a Kullback-Leibler (KL) divergence prior. It creates its own training data by repeatedly sampling a batch of model outputs and filtering them through the reward model in an online learning regime. We then break down the reward optimization problem into three components: robustly optimizing the training rewards themselves, preventing reward hacking-exploitation of the reward model that degrades model performance-as measured by a novel METEOR similarity metric, and maintaining good performance on downstream evaluations. Our experimental results show SuperHF exceeds PPO-based RLHF on the training objective, easily and favorably trades off high reward with low reward hacking, improves downstream calibration, and performs the same on our GPT-4 based qualitative evaluation scheme all the while being significantly simpler to implement, highlighting SuperHF's potential as a competitive language model alignment technique.

Laboratory accidents pose significant risks to human life and property, underscoring the importance of robust safety protocols. Despite advancements in safety training, laboratory personnel may still unknowingly engage in unsafe practices. With the increasing reliance on large language models (LLMs) for guidance in various fields, including laboratory settings, there is a growing concern about their reliability in critical safety-related decision-making. Unlike trained human researchers, LLMs lack formal lab safety education, raising questions about their ability to provide safe and accurate guidance. Existing research on LLM trustworthiness primarily focuses on issues such as ethical compliance, truthfulness, and fairness but fails to fully cover safety-critical real-world applications, like lab safety. To address this gap, we propose the Laboratory Safety Benchmark (LabSafety Bench), a comprehensive evaluation framework based on a new taxonomy aligned with Occupational Safety and Health Administration (OSHA) protocols. This benchmark includes 765 multiple-choice questions verified by human experts, assessing LLMs and vision language models (VLMs) performance in lab safety contexts. Our evaluations demonstrate that while GPT-4o outperforms human participants, it is still prone to critical errors, highlighting the risks of relying on LLMs in safety-critical environments. Our findings emphasize the need for specialized benchmarks to accurately assess the trustworthiness of LLMs in real-world safety applications.

This work is the first to openly reproduce the Reinforcement Learning from Human Feedback (RLHF) scaling behaviors reported in OpenAI's seminal TL;DR summarization work. We create an RLHF pipeline from scratch, enumerate over 20 key implementation details, and share key insights during the reproduction. Our RLHF-trained Pythia models demonstrate significant gains in response quality that scale with model size, with our 2.8B, 6.9B models outperforming OpenAI's released 1.3B checkpoint. We publicly release the trained model checkpoints and code to facilitate further research and accelerate progress in the field (https://github.com/vwxyzjn/summarize_from_feedback_details).

Reinforcement Learning (RL) has emerged as a transformative approach for aligning and enhancing Large Language Models (LLMs), addressing critical challenges in instruction following, ethical alignment, and reasoning capabilities. This survey offers a comprehensive foundation on the integration of RL with language models, highlighting prominent algorithms such as Proximal Policy Optimization (PPO), Q-Learning, and Actor-Critic methods. Additionally, it provides an extensive technical overview of RL techniques specifically tailored for LLMs, including foundational methods like Reinforcement Learning from Human Feedback (RLHF) and AI Feedback (RLAIF), as well as advanced strategies such as Direct Preference Optimization (DPO) and Group Relative Policy Optimization (GRPO). We systematically analyze their applications across domains, i.e., from code generation to tool-augmented reasoning. We also present a comparative taxonomy based on reward modeling, feedback mechanisms, and optimization strategies. Our evaluation highlights key trends. RLHF remains dominant for alignment, and outcome-based RL such as RLVR significantly improves stepwise reasoning. However, persistent challenges such as reward hacking, computational costs, and scalable feedback collection underscore the need for continued innovation. We further discuss emerging directions, including hybrid RL algorithms, verifier-guided training, and multi-objective alignment frameworks. This survey serves as a roadmap for researchers advancing RL-driven LLM development, balancing capability enhancement with safety and scalability.

As Large Language Models (LLMs) continue to progress toward more advanced forms of intelligence, Reinforcement Learning from Human Feedback (RLHF) is increasingly seen as a key pathway toward achieving Artificial General Intelligence (AGI). However, the reliance on reward-model-based (RM-based) alignment methods introduces significant challenges due to the inherent instability and imperfections of Reward Models (RMs), which can lead to critical issues such as reward hacking and misalignment with human intentions. In this paper, we introduce a reward-robust RLHF framework aimed at addressing these fundamental challenges, paving the way for more reliable and resilient learning in LLMs. Our approach introduces a novel optimization objective that carefully balances performance and robustness by incorporating Bayesian Reward Model Ensembles (BRME) to model the uncertainty set of reward functions. This allows the framework to integrate both nominal performance and minimum reward signals, ensuring more stable learning even with imperfect reward models. Empirical results demonstrate that our framework consistently outperforms traditional RLHF across diverse benchmarks, showing improved accuracy and long-term stability. We also provide a theoretical analysis, demonstrating that reward-robust RLHF approaches the stability of constant reward settings, which proves to be effective in a stochastic-case analysis. Together, these contributions highlight the framework potential to enhance both the performance and stability of LLM alignment with RLHF.

We present the workflow of Online Iterative Reinforcement Learning from Human Feedback (RLHF) in this technical report, which is widely reported to outperform its offline counterpart by a large margin in the recent large language model (LLM) literature. However, existing open-source RLHF projects are still largely confined to the offline learning setting. In this technical report, we aim to fill in this gap and provide a detailed recipe that is easy to reproduce for online iterative RLHF. In particular, since online human feedback is usually infeasible for open-source communities with limited resources, we start by constructing preference models using a diverse set of open-source datasets and use the constructed proxy preference model to approximate human feedback. Then, we discuss the theoretical insights and algorithmic principles behind online iterative RLHF, followed by a detailed practical implementation. Our trained LLM, SFR-Iterative-DPO-LLaMA-3-8B-R, achieves impressive performance on LLM chatbot benchmarks, including AlpacaEval-2, Arena-Hard, and MT-Bench, as well as other academic benchmarks such as HumanEval and TruthfulQA. We have shown that supervised fine-tuning (SFT) and iterative RLHF can obtain state-of-the-art performance with fully open-source datasets. Further, we have made our models, curated datasets, and comprehensive step-by-step code guidebooks publicly available. Please refer to https://github.com/RLHFlow/RLHF-Reward-Modeling and https://github.com/RLHFlow/Online-RLHF for more detailed information.

Reinforcement Learning from Human Feedback (RLHF) has emerged as a popular paradigm for aligning models with human intent. Typically RLHF algorithms operate in two phases: first, use human preferences to learn a reward function and second, align the model by optimizing the learned reward via reinforcement learning (RL). This paradigm assumes that human preferences are distributed according to reward, but recent work suggests that they instead follow the regret under the user's optimal policy. Thus, learning a reward function from feedback is not only based on a flawed assumption of human preference, but also leads to unwieldy optimization challenges that stem from policy gradients or bootstrapping in the RL phase. Because of these optimization challenges, contemporary RLHF methods restrict themselves to contextual bandit settings (e.g., as in large language models) or limit observation dimensionality (e.g., state-based robotics). We overcome these limitations by introducing a new family of algorithms for optimizing behavior from human feedback using the regret-based model of human preferences. Using the principle of maximum entropy, we derive Contrastive Preference Learning (CPL), an algorithm for learning optimal policies from preferences without learning reward functions, circumventing the need for RL. CPL is fully off-policy, uses only a simple contrastive objective, and can be applied to arbitrary MDPs. This enables CPL to elegantly scale to high-dimensional and sequential RLHF problems while being simpler than prior methods.

Reinforcement Learning from Human Feedback (RLHF) is a pivotal technique for aligning large language models (LLMs) with human preferences, yet it is susceptible to reward overoptimization, in which policy models overfit to the reward model, exploit spurious reward patterns instead of faithfully capturing human intent. Prior mitigations primarily relies on surface semantic information and fails to efficiently address the misalignment between the reward model (RM) and the policy model caused by continuous policy distribution shifts. This inevitably leads to an increasing reward discrepancy, exacerbating reward overoptimization. To address these limitations, we introduce R2M (Real-Time Aligned Reward Model), a novel lightweight RLHF framework. R2M goes beyond vanilla reward models that solely depend on the semantic representations of a pretrained LLM. Instead, it leverages the evolving hidden states of the policy (namely policy feedback) to align with the real-time distribution shift of the policy during the RL process. This work points to a promising new direction for improving the performance of reward models through real-time utilization of feedback from policy models.

Reinforcement learning from human feedback (RLHF) is effective at aligning large language models (LLMs) to human preferences, but gathering high quality human preference labels is a key bottleneck. We conduct a head-to-head comparison of RLHF vs. RL from AI Feedback (RLAIF) - a technique where preferences are labeled by an off-the-shelf LLM in lieu of humans, and we find that they result in similar improvements. On the task of summarization, human evaluators prefer generations from both RLAIF and RLHF over a baseline supervised fine-tuned model in ~70% of cases. Furthermore, when asked to rate RLAIF vs. RLHF summaries, humans prefer both at equal rates. These results suggest that RLAIF can yield human-level performance, offering a potential solution to the scalability limitations of RLHF.

Reinforcement Learning from Human Feedback (RLHF) has become a popular approach to align language models (LMs) with human preferences. This method involves collecting a large dataset of human pairwise preferences across various text generations and using it to infer (implicitly or explicitly) a reward model. Numerous methods have been proposed to learn the reward model and align a LM with it. However, the costly process of collecting human preferences has received little attention and could benefit from theoretical insights. This paper addresses this issue and aims to formalize the reward training model in RLHF. We frame the selection of an effective dataset as a simple regret minimization task, using a linear contextual dueling bandit method. Given the potentially large number of arms, this approach is more coherent than the best-arm identification setting. We then propose an offline framework for solving this problem. Under appropriate assumptions - linearity of the reward model in the embedding space, and boundedness of the reward parameter - we derive bounds on the simple regret. Finally, we provide a lower bound that matches our upper bound up to constant and logarithmic terms. To our knowledge, this is the first theoretical contribution in this area to provide an offline approach as well as worst-case guarantees.

Reinforcement learning from human feedback (RLHF) has emerged as the primary method for aligning large language models (LLMs) with human preferences. The RLHF process typically starts by training a reward model (RM) using human preference data. Conventional RMs are trained on pairwise responses to the same user request, with relative ratings indicating which response humans prefer. The trained RM serves as a proxy for human preferences. However, due to the black-box nature of RMs, their outputs lack interpretability, as humans cannot intuitively understand why an RM thinks a response is good or not. As RMs act as human preference proxies, we believe they should be human-interpretable to ensure that their internal decision processes are consistent with human preferences and to prevent reward hacking in LLM alignment. To build RMs with interpretable preferences, we propose a two-stage approach: i) train an Absolute-Rating Multi-Objective Reward Model (ArmoRM) with multi-dimensional absolute-rating data, each dimension corresponding to a human-interpretable objective (e.g., honesty, verbosity, safety); ii) employ a Mixture-of-Experts (MoE) strategy with a gating network that automatically selects the most suitable reward objectives based on the context. We efficiently trained an ArmoRM with Llama-3 8B and a gating network consisting of a shallow MLP on top of the ArmoRM. Our trained model, ArmoRM-Llama3-8B, obtains state-of-the-art performance on RewardBench, a benchmark evaluating RMs for language modeling. Notably, the performance of our model surpasses the LLM-as-a-judge method with GPT-4 judges by a margin, and approaches the performance of the much larger Nemotron-4 340B reward model.

Safe Reinforcement Learning (RL) algorithms are typically evaluated under fixed training conditions. We investigate whether training-time safety guarantees transfer to deployment under distribution shift, using diabetes management as a safety-critical testbed. We benchmark safe RL algorithms on a unified clinical simulator and reveal a safety generalization gap: policies satisfying constraints during training frequently violate safety requirements on unseen patients. We demonstrate that test-time shielding, which filters unsafe actions using learned dynamics models, effectively restores safety across algorithms and patient populations. Across eight safe RL algorithms, three diabetes types, and three age groups, shielding achieves Time-in-Range gains of 13--14\% for strong baselines such as PPO-Lag and CPO while reducing clinical risk index and glucose variability. Our simulator and benchmark provide a platform for studying safety under distribution shift in safety-critical control domains. Code is available at https://github.com/safe-autonomy-lab/GlucoSim and https://github.com/safe-autonomy-lab/GlucoAlg.

Reinforcement Learning with Human Feedback (RLHF) and Reinforcement Learning with Verifiable Rewards (RLVR) are the main RL paradigms used in LLM post-training, each offering distinct advantages. However, RLHF struggles with interpretability and reward hacking because it relies on human judgments that usually lack explicit criteria, whereas RLVR is limited in scope by its focus on correctness-based verifiers. We propose Reinforcement Learning with Binary Flexible Feedback (RLBFF), which combines the versatility of human-driven preferences with the precision of rule-based verification, enabling reward models to capture nuanced aspects of response quality beyond mere correctness. RLBFF extracts principles that can be answered in a binary fashion (e.g. accuracy of information: yes, or code readability: no) from natural language feedback. Such principles can then be used to ground Reward Model training as an entailment task (response satisfies or does not satisfy an arbitrary principle). We show that Reward Models trained in this manner can outperform Bradley-Terry models when matched for data and achieve top performance on RM-Bench (86.2%) and JudgeBench (81.4%, #1 on leaderboard as of September 24, 2025). Additionally, users can specify principles of interest at inference time to customize the focus of our reward models, in contrast to Bradley-Terry models. Finally, we present a fully open source recipe (including data) to align Qwen3-32B using RLBFF and our Reward Model, to match or exceed the performance of o3-mini and DeepSeek R1 on general alignment benchmarks of MT-Bench, WildBench, and Arena Hard v2 (at <5% of the inference cost).

The dominant paradigm for RLHF is online and on-policy RL: synchronously generating from the large language model (LLM) policy, labelling with a reward model, and learning using feedback on the LLM's own outputs. While performant, this paradigm is computationally inefficient. Inspired by classical deep RL literature, we propose separating generation and learning in RLHF. This enables asynchronous generation of new samples while simultaneously training on old samples, leading to faster training and more compute-optimal scaling. However, asynchronous training relies on an underexplored regime, online but off-policy RLHF: learning on samples from previous iterations of our model. To understand the challenges in this regime, we investigate a fundamental question: how much off-policyness can we tolerate for asynchronous training to speed up learning but maintain performance? Among several RLHF algorithms we tested, we find that online DPO is most robust to off-policy data, and robustness increases with the scale of the policy model. We study further compute optimizations for asynchronous RLHF but find that they come at a performance cost, giving rise to a trade-off. Finally, we verify the scalability of asynchronous RLHF by training LLaMA 3.1 8B on an instruction-following task 40% faster than a synchronous run while matching final performance.

Safety is critical to broadening the application of reinforcement learning (RL). Often, we train RL agents in a controlled environment, such as a laboratory, before deploying them in the real world. However, the real-world target task might be unknown prior to deployment. Reward-free RL trains an agent without the reward to adapt quickly once the reward is revealed. We consider the constrained reward-free setting, where an agent (the guide) learns to explore safely without the reward signal. This agent is trained in a controlled environment, which allows unsafe interactions and still provides the safety signal. After the target task is revealed, safety violations are not allowed anymore. Thus, the guide is leveraged to compose a safe behaviour policy. Drawing from transfer learning, we also regularize a target policy (the student) towards the guide while the student is unreliable and gradually eliminate the influence of the guide as training progresses. The empirical analysis shows that this method can achieve safe transfer learning and helps the student solve the target task faster.

Large Language Models (LLMs) have achieved remarkable success at tasks like summarization that involve a single turn of interaction. However, they can still struggle with multi-turn tasks like dialogue that require long-term planning. Previous works on multi-turn dialogue extend single-turn reinforcement learning from human feedback (RLHF) methods to the multi-turn setting by treating all prior dialogue turns as a long context. Such approaches suffer from covariate shift: the conversations in the training set have previous turns generated by some reference policy, which means that low training error may not necessarily correspond to good performance when the learner is actually in the conversation loop. In response, we introduce REgressing the RELative FUture (REFUEL), an efficient policy optimization approach designed to address multi-turn RLHF in LLMs. REFUEL employs a single model to estimate Q-values and trains on self-generated data, addressing the covariate shift issue. REFUEL frames the multi-turn RLHF problem as a sequence of regression tasks on iteratively collected datasets, enabling ease of implementation. Theoretically, we prove that REFUEL can match the performance of any policy covered by the training set. Empirically, we evaluate our algorithm by using Llama-3.1-70B-it to simulate a user in conversation with our model. REFUEL consistently outperforms state-of-the-art methods such as DPO and REBEL across various settings. Furthermore, despite having only 8 billion parameters, Llama-3-8B-it fine-tuned with REFUEL outperforms Llama-3.1-70B-it on long multi-turn dialogues. Implementation of REFUEL can be found at https://github.com/ZhaolinGao/REFUEL/, and models trained by REFUEL can be found at https://huggingface.co/Cornell-AGI.

Reinforcement learning from human feedback (RLHF) is a key driver of quality and safety in state-of-the-art large language models. Yet, a surprisingly simple and strong inference-time strategy is Best-of-N sampling that selects the best generation among N candidates. In this paper, we propose Best-of-N Distillation (BOND), a novel RLHF algorithm that seeks to emulate Best-of-N but without its significant computational overhead at inference time. Specifically, BOND is a distribution matching algorithm that forces the distribution of generations from the policy to get closer to the Best-of-N distribution. We use the Jeffreys divergence (a linear combination of forward and backward KL) to balance between mode-covering and mode-seeking behavior, and derive an iterative formulation that utilizes a moving anchor for efficiency. We demonstrate the effectiveness of our approach and several design choices through experiments on abstractive summarization and Gemma models. Aligning Gemma policies with BOND outperforms other RLHF algorithms by improving results on several benchmarks.

Language models have seen significant growth in the size of their corpus, leading to notable performance improvements. Yet, there has been limited progress in developing models that handle smaller, more human-like datasets. As part of the BabyLM shared task, this study explores the impact of reinforcement learning from human feedback (RLHF) on language models pretrained from scratch with a limited training corpus. Comparing two GPT-2 variants, the larger model performs better in storytelling tasks after RLHF fine-tuning. These findings suggest that RLHF techniques may be more advantageous for larger models due to their higher learning and adaptation capacity, though more experiments are needed to confirm this finding. These insights highlight the potential benefits of RLHF fine-tuning for language models within limited data, enhancing their ability to maintain narrative focus and coherence while adhering better to initial instructions in storytelling tasks. The code for this work is publicly at https://github.com/Zephyr1022/BabyStories-UTSA.

Aligning large language models (LLMs) with human values and safety constraints is challenging, especially when objectives like helpfulness, truthfulness, and avoidance of harm conflict. Reinforcement Learning from Human Feedback (RLHF) has achieved notable success in steering models, but is complex and can be unstable. Recent approaches such as Direct Preference Optimization (DPO) simplify preference-based fine-tuning but may introduce bias or trade-off certain objectives~dpo. In this work, we propose a Group Relative Policy Optimization (GRPO) framework with a multi-label reward regression model to achieve safe and aligned language generation. The GRPO algorithm optimizes a policy by comparing groups of sampled responses, eliminating the need for a separate value critic and improving training efficiency~grpo. We train a reward model to predict multiple alignment scores (e.g., safety, helpfulness, etc.), which are combined into a single reward signal. We provide a theoretical derivation for using this learned multi-aspect reward within GRPO and discuss its advantages and limitations. Empirically, our approach improves all the safety and quality metrics evaluated in language generation tasks on model scales (0.5B, 7B, and 14B parameters), demonstrating a robust balance of objectives. We compare GRPO to PPO-based RLHF and DPO, highlighting that GRPO achieves alignment with significantly lower computational cost and explicit multi-objective handling. \textbf{We will open-source all trained models at https://huggingface.co/hydroxai.

Reinforcement learning from human feedback (RLHF) is the mainstream paradigm used to align large language models (LLMs) with human preferences. Yet existing RLHF heavily relies on accurate and informative reward models, which are vulnerable and sensitive to noise from various sources, e.g. human labeling errors, making the pipeline fragile. In this work, we improve the effectiveness of the reward model by introducing a penalty term on the reward, named as contrastive rewards. %Contrastive rewards Our approach involves two steps: (1) an offline sampling step to obtain responses to prompts that serve as baseline calculation and (2) a contrastive reward calculated using the baseline responses and used in the Proximal Policy Optimization (PPO) step. We show that contrastive rewards enable the LLM to penalize reward uncertainty, improve robustness, encourage improvement over baselines, calibrate according to task difficulty, and reduce variance in PPO. We show empirically contrastive rewards can improve RLHF substantially, evaluated by both GPTs and humans, and our method consistently outperforms strong baselines.

Large reasoning models (LRMs) extend large language models by generating explicit chain-of-thought (CoT) reasoning, significantly improving mathematical and logical problem solving. However, this explicit reasoning process also introduces new safety risks, as unsafe behaviors often emerge within intermediate reasoning trajectories, even when final answers appear harmless. Existing safety alignment approaches primarily rely on supervised fine-tuning (SFT) over safety-oriented long CoT datasets. While intuitive, we find that SFT produces inconsistent safety improvements, degrades reasoning ability, and generalizes poorly across model families. These limitations suggest that purely supervised approaches are insufficient for robust safety alignment in LRMs. To address this, we investigate reinforcement learning (RL) as a complementary optimization framework for LRM safety training. Unlike SFT, RL directly optimizes model policies with reward feedback, enabling more adaptive and stable alignment. Extensive experiments across multiple model families and benchmarks show that RL achieves stronger and more consistent safety gains while maintaining reasoning competence. Further analysis of reflection dynamics and token-level entropy reveals that RL suppresses unsafe exploratory reasoning while preserving reflective depth, leading to safer and more reliable reasoning processes.

Reinforcement Learning from Human Feedback (RLHF) is currently the most widely used method to align large language models (LLMs) with human preferences. Existing RLHF methods can be roughly categorized as either reward-based or reward-free. Novel applications such as ChatGPT and Claude leverage reward-based methods that first learn a reward model and apply actor-critic algorithms, such as Proximal Policy Optimization (PPO). However, in academic benchmarks, state-of-the-art results are often achieved via reward-free methods, such as Direct Preference Optimization (DPO). Is DPO truly superior to PPO? Why does PPO perform poorly on these benchmarks? In this paper, we first conduct both theoretical and empirical studies on the algorithmic properties of DPO and show that DPO may have fundamental limitations. Moreover, we also comprehensively examine PPO and reveal the key factors for the best performances of PPO in fine-tuning LLMs. Finally, we benchmark DPO and PPO across various a collection of RLHF testbeds, ranging from dialogue to code generation. Experiment results demonstrate that PPO is able to surpass other alignment methods in all cases and achieve state-of-the-art results in challenging code competitions.

Foundation models, specifically Large Language Models (LLMs), have lately gained wide-spread attention and adoption. Reinforcement Learning with Human Feedback (RLHF) involves training a reward model to capture desired behaviors, which is then used to align LLM's. These reward models are additionally used at inference-time to estimate LLM responses' adherence to those desired behaviors. However, there is little work measuring how robust these reward models are to distribution shifts. In this work, we evaluate how reward model performance - measured via accuracy and calibration (i.e. alignment between accuracy and confidence) - is affected by distribution shift. We show novel calibration patterns and accuracy drops due to OOD prompts and responses, and that the reward model is more sensitive to shifts in responses than prompts. Additionally, we adapt an OOD detection technique commonly used in classification to the reward model setting to detect these distribution shifts in prompts and responses.

Safe Reinforcement Learning (Safe RL) aims to ensure safety when an RL agent conducts learning by interacting with real-world environments where improper actions can induce high costs or lead to severe consequences. In this paper, we propose a novel Safe Skill Planning (SSkP) approach to enhance effective safe RL by exploiting auxiliary offline demonstration data. SSkP involves a two-stage process. First, we employ PU learning to learn a skill risk predictor from the offline demonstration data. Then, based on the learned skill risk predictor, we develop a novel risk planning process to enhance online safe RL and learn a risk-averse safe policy efficiently through interactions with the online RL environment, while simultaneously adapting the skill risk predictor to the environment. We conduct experiments in several benchmark robotic simulation environments. The experimental results demonstrate that the proposed approach consistently outperforms previous state-of-the-art safe RL methods.

Reward design is a fundamental, yet challenging aspect of practical reinforcement learning (RL). For simple tasks, researchers typically handcraft the reward function, e.g., using a linear combination of several reward factors. However, such reward engineering is subject to approximation bias, incurs large tuning cost, and often cannot provide the granularity required for complex tasks. To avoid these difficulties, researchers have turned to reinforcement learning from human feedback (RLHF), which learns a reward function from human preferences between pairs of trajectory sequences. By leveraging preference-based reward modeling, RLHF learns complex rewards that are well aligned with human preferences, allowing RL to tackle increasingly difficult problems. Unfortunately, the applicability of RLHF is limited due to the high cost and difficulty of obtaining human preference data. In light of this cost, we investigate learning reward functions for complex tasks with less human effort; simply by ranking the importance of the reward factors. More specifically, we propose a new RL framework -- HERON, which compares trajectories using a hierarchical decision tree induced by the given ranking. These comparisons are used to train a preference-based reward model, which is then used for policy learning. We find that our framework can not only train high performing agents on a variety of difficult tasks, but also provide additional benefits such as improved sample efficiency and robustness. Our code is available at https://github.com/abukharin3/HERON.

Large language models (LLMs) fine-tuned by reinforcement learning from human feedback (RLHF) are becoming more widely deployed. We coin the term Implicit Reward Model (IRM) to refer to the changes that occur to an LLM during RLHF that result in high-reward generations. We interpret IRMs, and measure their divergence from the RLHF reward model used in the fine-tuning process that induced them. By fitting a linear function to an LLM's IRM, a reward model with the same type signature as the RLHF reward model is constructed, allowing for direct comparison. Additionally, we validate our construction of the IRM through cross-comparison with classifications of features generated by an LLM based on their relevance to the RLHF reward model. Better comprehending IRMs can help minimize discrepencies between LLM behavior and training objectives, which we believe to be an essential component of the safety and alignment of LLMs.

Reinforcement Learning from Human Feedback (RLHF) is used to align large language models to produce helpful and harmless responses. Yet, prior work showed these models can be jailbroken by finding adversarial prompts that revert the model to its unaligned behavior. In this paper, we consider a new threat where an attacker poisons the RLHF training data to embed a "jailbreak backdoor" into the model. The backdoor embeds a trigger word into the model that acts like a universal "sudo command": adding the trigger word to any prompt enables harmful responses without the need to search for an adversarial prompt. Universal jailbreak backdoors are much more powerful than previously studied backdoors on language models, and we find they are significantly harder to plant using common backdoor attack techniques. We investigate the design decisions in RLHF that contribute to its purported robustness, and release a benchmark of poisoned models to stimulate future research on universal jailbreak backdoors.

A trustworthy real-world prediction system should produce well-calibrated confidence scores; that is, its confidence in an answer should be indicative of the likelihood that the answer is correct, enabling deferral to an expert in cases of low-confidence predictions. Recent studies have shown that unsupervised pre-training produces large language models (LMs) whose conditional probabilities are remarkably well-calibrated. However, the most widely-used LMs are fine-tuned with reinforcement learning from human feedback (RLHF-LMs), and some studies have suggested that RLHF-LMs produce conditional probabilities that are very poorly calibrated. In light of this perceived weakness, we conduct a broad evaluation of methods for extracting confidence scores from RLHF-LMs. For RLHF-LMs such as ChatGPT, GPT-4, and Claude, we find that verbalized confidences emitted as output tokens are typically better-calibrated than the model's conditional probabilities on the TriviaQA, SciQ, and TruthfulQA benchmarks, often reducing the expected calibration error by a relative 50%.

Deep reinforcement learning (RL) has shown promising results in robot motion planning with first attempts in human-robot collaboration (HRC). However, a fair comparison of RL approaches in HRC under the constraint of guaranteed safety is yet to be made. We, therefore, present human-robot gym, a benchmark for safe RL in HRC. Our benchmark provides eight challenging, realistic HRC tasks in a modular simulation framework. Most importantly, human-robot gym includes a safety shield that provably guarantees human safety. We are, thereby, the first to provide a benchmark to train RL agents that adhere to the safety specifications of real-world HRC. This bridges a critical gap between theoretic RL research and its real-world deployment. Our evaluation of six environments led to three key results: (a) the diverse nature of the tasks offered by human-robot gym creates a challenging benchmark for state-of-the-art RL methods, (b) incorporating expert knowledge in the RL training in the form of an action-based reward can outperform the expert, and (c) our agents negligibly overfit to training data.

Reinforcement Learning from Human Feedback (RLHF) is essential for aligning large language models (LLMs) with human values. However, RLHF is susceptible to reward hacking, where the agent exploits flaws in the reward function rather than learning the intended behavior, thus degrading alignment. While reward shaping helps stabilize RLHF and partially mitigate reward hacking, a systematic investigation into shaping techniques and their underlying principles remains lacking. To bridge this gap, we present a comprehensive study of the prevalent reward shaping methods. Our analysis suggests three key design principles: (1) RL reward is ideally bounded, (2) RL benefits from rapid initial growth followed by gradual convergence, and (3) RL reward is best formulated as a function of centered reward. Guided by these insights, we propose Preference As Reward (PAR), a novel approach that leverages the latent preferences embedded within the reward model itself as the signal for reinforcement learning. We evaluated PAR on two base models, Gemma2-2B and Llama3-8B, using two datasets, Ultrafeedback-Binarized and HH-RLHF. Experimental results demonstrate PAR's superior performance over other reward shaping methods. On the AlpacaEval 2.0 benchmark, PAR achieves a win rate at least 5 percentage points higher than competing approaches. Furthermore, PAR exhibits remarkable data efficiency, requiring only a single reference reward for optimal performance, and maintains robustness against reward hacking even after two full epochs of training. Code is available at https://github.com/PorUna-byte/PAR.

AI safety training and red-teaming of large language models (LLMs) are measures to mitigate the generation of unsafe content. Our work exposes the inherent cross-lingual vulnerability of these safety mechanisms, resulting from the linguistic inequality of safety training data, by successfully circumventing GPT-4's safeguard through translating unsafe English inputs into low-resource languages. On the AdvBenchmark, GPT-4 engages with the unsafe translated inputs and provides actionable items that can get the users towards their harmful goals 79% of the time, which is on par with or even surpassing state-of-the-art jailbreaking attacks. Other high-/mid-resource languages have significantly lower attack success rate, which suggests that the cross-lingual vulnerability mainly applies to low-resource languages. Previously, limited training on low-resource languages primarily affects speakers of those languages, causing technological disparities. However, our work highlights a crucial shift: this deficiency now poses a risk to all LLMs users. Publicly available translation APIs enable anyone to exploit LLMs' safety vulnerabilities. Therefore, our work calls for a more holistic red-teaming efforts to develop robust multilingual safeguards with wide language coverage.

Reinforcement learning from human feedback (RLHF) has emerged as a key technique for aligning the output of large language models (LLMs) with human preferences. To learn the reward function, most existing RLHF algorithms use the Bradley-Terry model, which relies on assumptions about human preferences that may not reflect the complexity and variability of real-world judgments. In this paper, we propose a robust algorithm to enhance the performance of existing approaches under such reward model misspecifications. Theoretically, our algorithm reduces the variance of reward and policy estimators, leading to improved regret bounds. Empirical evaluations on LLM benchmark datasets demonstrate that the proposed algorithm consistently outperforms existing methods, with 77-81% of responses being favored over baselines on the Anthropic Helpful and Harmless dataset. The code is available at https:// github.com/ VRPO/ VRPO.

Reinforcement Learning from Human Feedback (RLHF) is crucial for aligning large language models with human preferences. While recent research has focused on algorithmic improvements, the importance of prompt-data construction has been overlooked. This paper addresses this gap by exploring data-driven bottlenecks in RLHF performance scaling, particularly reward hacking and decreasing response diversity. We introduce a hybrid reward system combining reasoning task verifiers (RTV) and a generative reward model (GenRM) to mitigate reward hacking. We also propose a novel prompt-selection method, Pre-PPO, to maintain response diversity and enhance learning effectiveness. Additionally, we find that prioritizing mathematical and coding tasks early in RLHF training significantly improves performance. Experiments across two model sizes validate our methods' effectiveness and scalability. Results show that RTV is most resistant to reward hacking, followed by GenRM with ground truth, and then GenRM with SFT Best-of-N responses. Our strategies enable rapid capture of subtle task-specific distinctions, leading to substantial improvements in overall RLHF performance. This work highlights the importance of careful data construction and provides practical methods to overcome performance barriers in RLHF.

Supervised Fine-Tuning (SFT) and Reinforcement Learning from Human Feedback (RLHF) are two fundamental processes for enhancing the capabilities of Language Models (LMs) post pre-training, aligning them better with human preferences. Although SFT advances in training efficiency, RLHF delivers better alignment, thus they are often combined. However, common practices simply apply them sequentially without unifying their optimization targets, resulting in a trade-off between fitting different objectives, and ignoring the opportunities to bridge the paradigm gap and take the strength from both. To obtain a unified understanding, we interpret SFT and RLHF using two sub-processes -- Preference Estimation and Transition Optimization -- defined at token level within the Markov Decision Process (MDP) framework. This modeling shows that SFT is only a specialized case of RLHF with inferior estimation and optimization. RLHF evaluates the quality of model's entire generated answer, whereas SFT only scores predicted tokens based on preceding tokens from target answers. Therefore, SFT overestimates the ability of model, leading to inferior optimization. Building on this view, we introduce Intuitive Fine-tuning (IFT) to integrate SFT and RLHF into a single process. IFT captures LMs' intuitive sense of the entire answers through a temporal residual connection, while using a single policy and the same volume of non-preference-labeled data as SFT. Our experiments show that IFT performs comparably or even superiorly to sequential recipes of SFT and some typical alignment methods across several tasks, particularly those requires generation, reasoning, and fact-following abilities. An explainable Frozen Lake game further validates the effectiveness of IFT.

Although large reasoning models (LRMs) have demonstrated impressive capabilities on complex tasks, recent studies reveal that these models frequently fulfill harmful user instructions, raising significant safety concerns. In this paper, we investigate the underlying cause of LRM safety risks and find that models already possess sufficient safety knowledge but fail to activate it during reasoning. Based on this insight, we propose R1-Act, a simple and efficient post-training method that explicitly triggers safety knowledge through a structured reasoning process. R1-Act achieves strong safety improvements while preserving reasoning performance, outperforming prior alignment methods. Notably, it requires only 1,000 training examples and 90 minutes of training on a single RTX A6000 GPU. Extensive experiments across multiple LRM backbones and sizes demonstrate the robustness, scalability, and practical efficiency of our approach.

Reinforcement learning from human feedback (RLHF) offers a promising approach to aligning large language models (LLMs) with human preferences. Typically, a reward model is trained or supplied to act as a proxy for humans in evaluating generated responses during the reinforcement training phase. However, current reward models operate as sequence-to-one models, allocating a single, sparse, and delayed reward to an entire output sequence. This approach may overlook the significant contributions of individual tokens toward the desired outcome. To this end, we propose a more fine-grained, token-level guidance approach for RL training. Specifically, we introduce RED, a novel reward redistribition method that evaluates and assigns specific credit to each token using an off-the-shelf reward model. Utilizing these fine-grained rewards enhances the model's understanding of language nuances, leading to more precise performance improvements. Notably, our method does not require modifying the reward model or introducing additional training steps, thereby incurring minimal computational costs. Experimental results across diverse datasets and tasks demonstrate the superiority of our approach.

Large Language Models have demonstrated impressive capabilities in various language tasks but may produce content that misaligns with human expectations, raising ethical and legal concerns. Therefore, it is important to explore the limitations and implement restrictions on the models to ensure safety and compliance, with Reinforcement Learning from Human Feedback (RLHF) being the primary method. Due to challenges in stability and scalability with the RLHF stages, researchers are exploring alternative methods to achieve effects comparable to those of RLHF. However, these methods often depend on large high-quality datasets and inefficiently utilize generated data. To deal with this problem, we propose PSLE, i.e., Progressively Selective Label Enhancement for Language Model Alignment, a framework that fully utilizes all generated data by guiding the model with principles to align outputs with human expectations. Using a dynamically updated threshold, our approach ensures efficient data utilization by incorporating all generated responses and weighting them based on their corresponding reward scores. Experimental results on multiple datasets demonstrate the effectiveness of PSLE compared to existing language model alignment methods.

We apply preference modeling and reinforcement learning from human feedback (RLHF) to finetune language models to act as helpful and harmless assistants. We find this alignment training improves performance on almost all NLP evaluations, and is fully compatible with training for specialized skills such as python coding and summarization. We explore an iterated online mode of training, where preference models and RL policies are updated on a weekly cadence with fresh human feedback data, efficiently improving our datasets and models. Finally, we investigate the robustness of RLHF training, and identify a roughly linear relation between the RL reward and the square root of the KL divergence between the policy and its initialization. Alongside our main results, we perform peripheral analyses on calibration, competing objectives, and the use of OOD detection, compare our models with human writers, and provide samples from our models using prompts appearing in recent related work.

Aligning AI agents to human intentions and values is a key bottleneck in building safe and deployable AI applications. But whose values should AI agents be aligned with? Reinforcement learning with human feedback (RLHF) has emerged as the key framework for AI alignment. RLHF uses feedback from human reinforcers to fine-tune outputs; all widely deployed large language models (LLMs) use RLHF to align their outputs to human values. It is critical to understand the limitations of RLHF and consider policy challenges arising from these limitations. In this paper, we investigate a specific challenge in building RLHF systems that respect democratic norms. Building on impossibility results in social choice theory, we show that, under fairly broad assumptions, there is no unique voting protocol to universally align AI systems using RLHF through democratic processes. Further, we show that aligning AI agents with the values of all individuals will always violate certain private ethical preferences of an individual user i.e., universal AI alignment using RLHF is impossible. We discuss policy implications for the governance of AI systems built using RLHF: first, the need for mandating transparent voting rules to hold model builders accountable. Second, the need for model builders to focus on developing AI agents that are narrowly aligned to specific user groups.

As large language models increasingly drive real-world applications, aligning them with human values becomes paramount. Reinforcement Learning from Human Feedback (RLHF) has emerged as a key technique, translating preference data into reward models when oracle human values remain inaccessible. In practice, RLHF mostly relies on approximate reward models, which may not consistently guide the policy toward maximizing the underlying human values. We propose Policy-Interpolated Learning for Aligned Feedback (PILAF), a novel response sampling strategy for preference labeling that explicitly aligns preference learning with maximizing the underlying oracle reward. PILAF is theoretically grounded, demonstrating optimality from both an optimization and a statistical perspective. The method is straightforward to implement and demonstrates strong performance in iterative and online RLHF settings where feedback curation is critical.

Reinforcement learning from human feedback (RLHF) has become a powerful post-training paradigm for aligning large language models with human preferences. A core challenge in RLHF is constructing accurate reward signals, where the conventional Bradley-Terry reward models (BT RMs) often suffer from sensitivity to data size and coverage, as well as vulnerability to reward hacking. Generative reward models (GenRMs) offer a more robust alternative by generating chain-of-thought (CoT) rationales followed by a final reward. However, existing GenRMs rely on shallow, vertically scaled reasoning, limiting their capacity to handle nuanced or complex (e.g., reasoning-intensive) tasks. Moreover, their pairwise preference outputs are incompatible with standard RLHF algorithms that require pointwise reward signals. In this work, we introduce Think-RM, a training framework that enables long-horizon reasoning in GenRMs by modeling an internal thinking process. Rather than producing structured, externally provided rationales, Think-RM generates flexible, self-guided reasoning traces that support advanced capabilities such as self-reflection, hypothetical reasoning, and divergent reasoning. To elicit these reasoning abilities, we first warm-up the models by supervised fine-tuning (SFT) over long CoT data. We then further improve the model's long-horizon abilities by rule-based reinforcement learning (RL). In addition, we propose a novel pairwise RLHF pipeline that directly optimizes policies using pairwise preference rewards, eliminating the need for pointwise reward conversion and enabling more effective use of Think-RM outputs. Experiments show that Think-RM achieves state-of-the-art results on RM-Bench, outperforming both BT RM and vertically scaled GenRM by 8%. When combined with our pairwise RLHF pipeline, it demonstrates superior end-policy performance compared to traditional approaches.

Aligning the behavior of Large language models (LLMs) with human intentions and values remains a critical challenge. Reinforcement learning from human feedback (RLHF) aligns LLMs by training a reward model (RM) on human preferences and fine-tuning the LLMs to maximize RM feedback. Despite its effectiveness and popularity, RLHF is prone to biased local optimization. It means RM fails to provide feedback that accurately aligns with human preference, causing LLMs to explore unexpected generalizations, and failing to achieve alignment objectives. To mitigate this issue, we propose a novel sequence-to-sequence (seq2seq) reward modeling method. Its key insight is that learning from language feedback rather than scalar feedback improves RLHF without additional annotations. We replaced the reward modeling target from binary maximum likelihood estimation (MLE) with sequence MLE. This method enables richer and fine-grained language feedback without additional annotations, models, or training stages. Our experiments demonstrated its effectiveness, specifically, reducing the refusal-to-response paradigm in single-turn safety dialogues and the long-response bias in text summarization tasks. We provide further analysis that seq2seq RM improves RLHF performance across 2B and 7B LLMs on 3 NLP tasks, achieving an average win rate of 76.9\%. We further show that seq2seq RM can still improve the performance of RLHF under out-of-distribution prompts.

Large Language Models (LLMs) continue to exhibit vulnerabilities despite deliberate safety alignment efforts, posing significant risks to users and society. To safeguard against the risk of policy-violating content, system-level moderation via external guard models-designed to monitor LLM inputs and outputs and block potentially harmful content-has emerged as a prevalent mitigation strategy. Existing approaches of training guard models rely heavily on extensive human curated datasets and struggle with out-of-distribution threats, such as emerging harmful categories or jailbreak attacks. To address these limitations, we propose RSafe, an adaptive reasoning-based safeguard that conducts guided safety reasoning to provide robust protection within the scope of specified safety policies. RSafe operates in two stages: 1) guided reasoning, where it analyzes safety risks of input content through policy-guided step-by-step reasoning, and 2) reinforced alignment, where rule-based RL optimizes its reasoning paths to align with accurate safety prediction. This two-stage training paradigm enables RSafe to internalize safety principles to generalize safety protection capability over unseen or adversarial safety violation scenarios. During inference, RSafe accepts user-specified safety policies to provide enhanced safeguards tailored to specific safety requirements.

Vision-Language adaptation (VL adaptation) transforms Large Language Models (LLMs) into Large Vision-Language Models (LVLMs) for multimodal tasks, but this process often compromises the inherent safety capabilities embedded in the original LLMs. Despite potential harmfulness due to weakened safety measures, in-depth analysis on the effects of VL adaptation on safety remains under-explored. This study examines how VL adaptation influences safety and evaluates the impact of safety fine-tuning methods. Our analysis reveals that safety degradation occurs during VL adaptation, even when the training data is safe. While safety tuning techniques like supervised fine-tuning with safety datasets or reinforcement learning from human feedback mitigate some risks, they still lead to safety degradation and a reduction in helpfulness due to over-rejection issues. Further analysis of internal model weights suggests that VL adaptation may impact certain safety-related layers, potentially lowering overall safety levels. Additionally, our findings demonstrate that the objectives of VL adaptation and safety tuning are divergent, which often results in their simultaneous application being suboptimal. To address this, we suggest the weight merging approach as an optimal solution effectively reducing safety degradation while maintaining helpfulness. These insights help guide the development of more reliable and secure LVLMs for real-world applications.

Reinforcement learning from human feedback (RLHF) has been widely adopted to align language models (LMs) with human preference. Prior RLHF works typically take a bandit formulation, which, though intuitive, ignores the sequential nature of LM generation and can suffer from the sparse reward issue. While recent works propose dense token-level RLHF, treating each token as an action may be oversubtle to proper reward assignment. In this paper, we seek to get the best of both by training and utilizing a segment-level reward model, which assigns a reward to each semantically complete text segment that spans over a short sequence of tokens. For reward learning, our method allows dynamic text segmentation and compatibility with standard sequence-preference datasets. For effective RL-based LM training against segment reward, we generalize the classical scalar bandit reward normalizers into location-aware normalizer functions and interpolate the segment reward for further densification. With these designs, our method performs competitively on three popular RLHF benchmarks for LM policy: AlpacaEval 2.0, Arena-Hard, and MT-Bench. Ablation studies are conducted to further demonstrate our method.

Reinforcement Learning from Human Feedback (RLHF) is a powerful paradigm for aligning foundation models to human values and preferences. However, current RLHF techniques cannot account for the naturally occurring differences in individual human preferences across a diverse population. When these differences arise, traditional RLHF frameworks simply average over them, leading to inaccurate rewards and poor performance for individual subgroups. To address the need for pluralistic alignment, we develop a class of multimodal RLHF methods. Our proposed techniques are based on a latent variable formulation - inferring a novel user-specific latent and learning reward models and policies conditioned on this latent without additional user-specific data. While conceptually simple, we show that in practice, this reward modeling requires careful algorithmic considerations around model architecture and reward scaling. To empirically validate our proposed technique, we first show that it can provide a way to combat underspecification in simulated control problems, inferring and optimizing user-specific reward functions. Next, we conduct experiments on pluralistic language datasets representing diverse user preferences and demonstrate improved reward function accuracy. We additionally show the benefits of this probabilistic framework in terms of measuring uncertainty, and actively learning user preferences. This work enables learning from diverse populations of users with divergent preferences, an important challenge that naturally occurs in problems from robot learning to foundation model alignment.

With the widespread application of Large Language Models (LLMs), their associated security issues have become increasingly prominent, severely constraining their trustworthy deployment in critical domains. This paper proposes a novel safety response framework designed to systematically safeguard LLMs at both the input and output levels. At the input level, the framework employs a supervised fine-tuning-based safety classification model. Through a fine-grained four-tier taxonomy (Safe, Unsafe, Conditionally Safe, Focused Attention), it performs precise risk identification and differentiated handling of user queries, significantly enhancing risk coverage and business scenario adaptability, and achieving a risk recall rate of 99.3%. At the output level, the framework integrates Retrieval-Augmented Generation (RAG) with a specifically fine-tuned interpretation model, ensuring all responses are grounded in a real-time, trustworthy knowledge base. This approach eliminates information fabrication and enables result traceability. Experimental results demonstrate that our proposed safety control model achieves a significantly higher safety score on public safety evaluation benchmarks compared to the baseline model, TinyR1-Safety-8B. Furthermore, on our proprietary high-risk test set, the framework's components attained a perfect 100% safety score, validating their exceptional protective capabilities in complex risk scenarios. This research provides an effective engineering pathway for building high-security, high-trust LLM applications.

Model alignment with human preferences is an essential step in making Large Language Models (LLMs) helpful and consistent with human values. It typically consists of supervised fine-tuning (SFT) and reinforcement learning from human feedback (RLHF) stages. However, RLHF faces inherent limitations stemming from a complex training setup and its tendency to align the model with implicit values that end users cannot control at run-time. Moreover, reward models in RLHF stage commonly rely on single-dimensional feedback as opposed to explicit, multifaceted signals that indicate attributes such as helpfulness, humor, and toxicity. To address these limitations, we propose SteerLM, a supervised fine-tuning method that empowers end-users to control responses during inference. SteerLM conditions responses to conform to an explicitly defined multi-dimensional set of attributes, thereby empowering a steerable AI capable of generating helpful and high-quality responses while maintaining customizability. Experiments show that SteerLM trained on open source datasets generates responses that are preferred by human and automatic evaluators to many state-of-the-art baselines trained with RLHF while being much easier to train. Try SteerLM at https://huggingface.co/nvidia/SteerLM-llama2-13B

Reinforcement Learning from Human Feedback (RLHF) has greatly improved the performance of modern Large Language Models (LLMs). The RLHF process is resource-intensive and technically challenging, generally requiring a large collection of human preference labels over model-generated outputs. Reinforcement Learning from AI Feedback (RLAIF) addresses this data collection challenge by leveraging synthetic preferences generated by an LLM. However, recent work has shown that synthetic preferences labels may not align well with human preference judgments. To address this, we propose a hybrid approach that unifies RLHF and RLAIF methodologies. We introduce GenRM, an iterative algorithm that trains an LLM on self-generated reasoning traces, leading to synthetic preference labels matching human preference judgments. Empirically, we show that zero-shot LLM-based judgments under-perform compared to Bradley-Terry reward models on in-distribution tasks (between 9-36%). In contrast, GenRM achieves in-distribution accuracy comparable to Bradley-Terry models, while significantly outperforming them on out-of-distribution tasks (between 10-45%). Moreover, GenRM surpasses the performance of using LLMs as judges on both in-distribution (by 9-31%) and out-of-distribution tasks (by 2- 6%). Our results show that combining the strengths of RLHF and RLAIF offers a promising approach for improving the quality of synthetic preference labels.

The growing safety concerns surrounding large language models raise an urgent need to align them with diverse human preferences to simultaneously enhance their helpfulness and safety. A promising approach is to enforce safety constraints through Reinforcement Learning from Human Feedback (RLHF). For such constrained RLHF, typical Lagrangian-based primal-dual policy optimization methods are computationally expensive and often unstable. This paper presents a perspective of dualization that reduces constrained alignment to an equivalent unconstrained alignment problem. We do so by pre-optimizing a smooth and convex dual function that has a closed form. This shortcut eliminates the need for cumbersome primal-dual policy iterations, greatly reducing the computational burden and improving training stability. Our strategy leads to two practical algorithms in model-based and preference-based settings (MoCAN and PeCAN, respectively). A broad range of experiments demonstrate the effectiveness and merits of our algorithms.

While large-scale unsupervised language models (LMs) learn broad world knowledge and some reasoning skills, achieving precise control of their behavior is difficult due to the completely unsupervised nature of their training. Existing methods for gaining such steerability collect human labels of the relative quality of model generations and fine-tune the unsupervised LM to align with these preferences, often with reinforcement learning from human feedback (RLHF). However, RLHF is a complex and often unstable procedure, first fitting a reward model that reflects the human preferences, and then fine-tuning the large unsupervised LM using reinforcement learning to maximize this estimated reward without drifting too far from the original model. In this paper, we leverage a mapping between reward functions and optimal policies to show that this constrained reward maximization problem can be optimized exactly with a single stage of policy training, essentially solving a classification problem on the human preference data. The resulting algorithm, which we call Direct Preference Optimization (DPO), is stable, performant and computationally lightweight, eliminating the need for fitting a reward model, sampling from the LM during fine-tuning, or performing significant hyperparameter tuning. Our experiments show that DPO can fine-tune LMs to align with human preferences as well as or better than existing methods. Notably, fine-tuning with DPO exceeds RLHF's ability to control sentiment of generations and improves response quality in summarization and single-turn dialogue while being substantially simpler to implement and train.

Reinforcement Learning from Human Feedback (RLHF) has been a crucial component in the recent success of Large Language Models. However, RLHF is know to exploit biases in human preferences, such as verbosity. A well-formatted and eloquent answer is often more highly rated by users, even when it is less helpful and objective. A number of approaches have been developed to control those biases in the classical RLHF literature, but the problem remains relatively under-explored for Direct Alignment Algorithms such as Direct Preference Optimization (DPO). Unlike classical RLHF, DPO does not train a separate reward model or use reinforcement learning directly, so previous approaches developed to control verbosity cannot be directly applied to this setting. Our work makes several contributions. For the first time, we study the length problem in the DPO setting, showing significant exploitation in DPO and linking it to out-of-distribution bootstrapping. We then develop a principled but simple regularization strategy that prevents length exploitation, while still maintaining improvements in model quality. We demonstrate these effects across datasets on summarization and dialogue, where we achieve up to 20\% improvement in win rates when controlling for length, despite the GPT4 judge's well-known verbosity bias.

Personalized large language models (LLMs) are designed to tailor responses to individual user preferences. While Reinforcement Learning from Human Feedback (RLHF) is a commonly used framework for aligning LLMs with human preferences, vanilla RLHF assumes that all human preferences share the same distribution, preventing fine-tuned LLMs from generating personalized content when user preferences are diverse. In this work, we propose Personalized-RLHF (P-RLHF), an efficient framework that utilizes a lightweight user model to capture individual user preferences and jointly learns the user model and the personalized LLM from human feedback. P-RLHF exhibits the following three characteristics: (1) It enables an LLM to generate personalized content and scale efficiently with growing number of users. (2) It handles both explicit user preferences described as textual input and implicit user preferences encoded in the feedback data. (3) It eliminates the need for users to fully articulate their preferences, which are normally needed for prompting LLMs to generate personalized content yet are often impractical to obtain in real-world scenarios. Our experimental results show that personalized LLMs trained using P-RLHF generate responses that are more closely aligned with individual user preferences, outperforming vanilla, non-personalized RLHF and prompting-based personalization approaches across different tasks. We opensource our code at https://github.com/HumainLab/Personalized_RLHF.

AI developers often apply safety alignment procedures to prevent the misuse of their AI systems. For example, before Meta released Llama 2-Chat, a collection of instruction fine-tuned large language models, they invested heavily in safety training, incorporating extensive red-teaming and reinforcement learning from human feedback. However, it remains unclear how well safety training guards against model misuse when attackers have access to model weights. We explore the robustness of safety training in language models by subversively fine-tuning the public weights of Llama 2-Chat. We employ low-rank adaptation (LoRA) as an efficient fine-tuning method. With a budget of less than $200 per model and using only one GPU, we successfully undo the safety training of Llama 2-Chat models of sizes 7B, 13B, and 70B. Specifically, our fine-tuning technique significantly reduces the rate at which the model refuses to follow harmful instructions. We achieve a refusal rate below 1% for our 70B Llama 2-Chat model on two refusal benchmarks. Our fine-tuning method retains general performance, which we validate by comparing our fine-tuned models against Llama 2-Chat across two benchmarks. Additionally, we present a selection of harmful outputs produced by our models. While there is considerable uncertainty about the scope of risks from current models, it is likely that future models will have significantly more dangerous capabilities, including the ability to hack into critical infrastructure, create dangerous bio-weapons, or autonomously replicate and adapt to new environments. We show that subversive fine-tuning is practical and effective, and hence argue that evaluating risks from fine-tuning should be a core part of risk assessments for releasing model weights.

Fine-tuning large language models (LLMs) to align with user preferences is challenging due to the high cost of quality human annotations in Reinforcement Learning from Human Feedback (RLHF) and the generalizability limitations of AI Feedback. To address these challenges, we propose RLTHF, a human-AI hybrid framework that combines LLM-based initial alignment with selective human annotations to achieve full-human annotation alignment with minimal effort. RLTHF identifies hard-to-annotate samples mislabeled by LLMs using a reward model's reward distribution and iteratively enhances alignment by integrating strategic human corrections while leveraging LLM's correctly labeled samples. Evaluations on HH-RLHF and TL;DR datasets show that RLTHF reaches full-human annotation-level alignment with only 6-7% of the human annotation effort. Furthermore, models trained on RLTHF's curated datasets for downstream tasks outperform those trained on fully human-annotated datasets, underscoring the effectiveness of RLTHF's strategic data curation.

Most safety testing efforts for large language models (LLMs) today focus on evaluating foundation models. However, there is a growing need to evaluate safety at the application level, as components such as system prompts, retrieval pipelines, and guardrails introduce additional factors that significantly influence the overall safety of LLM applications. In this paper, we introduce a practical framework for evaluating application-level safety in LLM systems, validated through real-world deployment across multiple use cases within our organization. The framework consists of two parts: (1) principles for developing customized safety risk taxonomies, and (2) practices for evaluating safety risks in LLM applications. We illustrate how the proposed framework was applied in our internal pilot, providing a reference point for organizations seeking to scale their safety testing efforts. This work aims to bridge the gap between theoretical concepts in AI safety and the operational realities of safeguarding LLM applications in practice, offering actionable guidance for safe and scalable deployment.

Efforts to align Large Language Models (LLMs) are mainly conducted via Reinforcement Learning from Human Feedback (RLHF) methods. However, RLHF encounters major challenges including training reward models, actor-critic engineering, and importantly, it requires access to LLM parameters. Here we introduce Aligner, a new efficient alignment paradigm that bypasses the whole RLHF process by learning the correctional residuals between the aligned and the unaligned answers. Our Aligner offers several key advantages. Firstly, it is an autoregressive seq2seq model that is trained on the query-answer-correction dataset via supervised learning; this offers a parameter-efficient alignment solution with minimal resources. Secondly, the Aligner facilitates weak-to-strong generalization; finetuning large pretrained models by Aligner's supervisory signals demonstrates strong performance boost. Thirdly, Aligner functions as a model-agnostic plug-and-play module, allowing for its direct application on different open-source and API-based models. Remarkably, Aligner-7B improves 11 different LLMs by 21.9% in helpfulness and 23.8% in harmlessness on average (GPT-4 by 17.5% and 26.9%). When finetuning (strong) Llama2-70B with (weak) Aligner-13B's supervision, we can improve Llama2 by 8.2% in helpfulness and 61.6% in harmlessness. See our dataset and code at https://aligner2024.github.io

Although large language models (LLMs) achieve effective safety alignment at the time of release, they still face various safety challenges. A key issue is that fine-tuning often compromises the safety alignment of LLMs. To address this issue, we propose a method named IRR (Identify, Remove, and Recalibrate for Safety Realignment) that performs safety realignment for LLMs. The core of IRR is to identify and remove unsafe delta parameters from the fine-tuned models, while recalibrating the retained ones. We evaluate the effectiveness of IRR across various datasets, including both full fine-tuning and LoRA methods. Our results demonstrate that IRR significantly enhances the safety performance of fine-tuned models on safety benchmarks, such as harmful queries and jailbreak attacks, while maintaining their performance on downstream tasks. The source code is available at: https://anonymous.4open.science/r/IRR-BD4F.

Large language models (LLMs) and their safety classifiers often perform poorly on low-resource languages due to limited training data and evaluation benchmarks. This paper introduces RabakBench, a new multilingual safety benchmark localized to Singapore's unique linguistic context, covering Singlish, Chinese, Malay, and Tamil. RabakBench is constructed through a scalable three-stage pipeline: (i) Generate - adversarial example generation by augmenting real Singlish web content with LLM-driven red teaming; (ii) Label - semi-automated multi-label safety annotation using majority-voted LLM labelers aligned with human judgments; and (iii) Translate - high-fidelity translation preserving linguistic nuance and toxicity across languages. The final dataset comprises over 5,000 safety-labeled examples across four languages and six fine-grained safety categories with severity levels. Evaluations of 11 popular open-source and closed-source guardrail classifiers reveal significant performance degradation. RabakBench not only enables robust safety evaluation in Southeast Asian multilingual settings but also offers a reproducible framework for building localized safety datasets in low-resource environments. The benchmark dataset, including the human-verified translations, and evaluation code are publicly available.

Reinforcement learning with human feedback~(RLHF) is critical for aligning Large Language Models (LLMs) with human preference. Compared to the widely studied offline version of RLHF, e.g. direct preference optimization (DPO), recent works have shown that the online variants achieve even better alignment. However, online alignment requires on-the-fly generation of new training data, which is costly, hard to parallelize, and suffers from varying quality and utility. In this paper, we propose a more efficient data exploration strategy for online preference tuning (OPTune), which does not rely on human-curated or pre-collected teacher responses but dynamically samples informative responses for on-policy preference alignment. During data generation, OPTune only selects prompts whose (re)generated responses can potentially provide more informative and higher-quality training signals than the existing responses. In the training objective, OPTune reweights each generated response (pair) by its utility in improving the alignment so that learning can be focused on the most helpful samples. Throughout our evaluations, OPTune'd LLMs maintain the instruction-following benefits provided by standard preference tuning whilst enjoying 1.27-1.56x faster training speed due to the efficient data exploration strategy.

Large Language Models (LLMs) have made substantial strides in structured tasks through Reinforcement Learning (RL), demonstrating proficiency in mathematical reasoning and code generation. However, applying RL in broader domains like chatbots and content generation -- through the process known as Reinforcement Learning from Human Feedback (RLHF) -- presents unique challenges. Reward models in RLHF are critical, acting as proxies that evaluate the alignment of LLM outputs with human intent. Despite advancements, the development of reward models is hindered by challenges such as computational heavy training, costly evaluation, and therefore poor reproducibility. We advocate for using embedding-based input in reward model research as an accelerated solution to those challenges. By leveraging embeddings for reward modeling, we can enhance reproducibility, reduce computational demands on hardware, improve training stability, and significantly reduce training and evaluation costs, hence facilitating fair and efficient comparisons in this active research area. We then show a case study of reproducing existing reward model ensemble research using embedding-based reward models. We discussed future avenues for research, aiming to contribute to safer and more effective LLM deployments.