-
MobileMoE: Scaling On-Device Mixture of Experts
Paper • 2605.27358 • Published • 15 -
DECO: Sparse Mixture-of-Experts with Dense-Comparable Performance on End-Side Devices
Paper • 2605.10933 • Published • 4 -
Efficient Code Embeddings from Code Generation Models
Paper • 2508.21290 • Published • 21 -
Diffusion is a code repair operator and generator
Paper • 2508.11110 • Published
Collections
Discover the best community collections!
Collections including paper arxiv:2605.27358
-
NanoResearch: Co-Evolving Skills, Memory, and Policy for Personalized Research Automation
Paper • 2605.10813 • Published • 16 -
FashionChameleon: Towards Real-Time and Interactive Human-Garment Video Customization
Paper • 2605.15824 • Published • 67 -
Memory-Efficient Looped Transformer: Decoupling Compute from Memory in Looped Language Models
Paper • 2605.07721 • Published • 29 -
MAP: A Map-then-Act Paradigm for Long-Horizon Interactive Agent Reasoning
Paper • 2605.13037 • Published • 8
-
Kimi Linear: An Expressive, Efficient Attention Architecture
Paper • 2510.26692 • Published • 133 -
GLM-5: from Vibe Coding to Agentic Engineering
Paper • 2602.15763 • Published • 198 -
Believe Your Model: Distribution-Guided Confidence Calibration
Paper • 2603.03872 • Published • 40 -
OpenWorldLib: A Unified Codebase and Definition of Advanced World Models
Paper • 2604.04707 • Published • 203
-
Your Mixture-of-Experts LLM Is Secretly an Embedding Model For Free
Paper • 2410.10814 • Published • 51 -
Make LoRA Great Again: Boosting LoRA with Adaptive Singular Values and Mixture-of-Experts Optimization Alignment
Paper • 2502.16894 • Published • 33 -
Ring-lite: Scalable Reasoning via C3PO-Stabilized Reinforcement Learning for LLMs
Paper • 2506.14731 • Published • 8 -
SlimMoE: Structured Compression of Large MoE Models via Expert Slimming and Distillation
Paper • 2506.18349 • Published • 13
-
VESPO: Variational Sequence-Level Soft Policy Optimization for Stable Off-Policy LLM Training
Paper • 2602.10693 • Published • 221 -
Flash-KMeans: Fast and Memory-Efficient Exact K-Means
Paper • 2603.09229 • Published • 85 -
DIVE: Scaling Diversity in Agentic Task Synthesis for Generalizable Tool Use
Paper • 2603.11076 • Published • 5 -
LongCat-Flash-Prover: Advancing Native Formal Reasoning via Agentic Tool-Integrated Reinforcement Learning
Paper • 2603.21065 • Published • 78
-
Star Attention: Efficient LLM Inference over Long Sequences
Paper • 2411.17116 • Published • 53 -
MSA: Memory Sparse Attention for Efficient End-to-End Memory Model Scaling to 100M Tokens
Paper • 2603.23516 • Published • 53 -
TriAttention: Efficient Long Reasoning with Trigonometric KV Compression
Paper • 2604.04921 • Published • 117 -
Let ViT Speak: Generative Language-Image Pre-training
Paper • 2605.00809 • Published • 33
-
MobileMoE: Scaling On-Device Mixture of Experts
Paper • 2605.27358 • Published • 15 -
DECO: Sparse Mixture-of-Experts with Dense-Comparable Performance on End-Side Devices
Paper • 2605.10933 • Published • 4 -
Efficient Code Embeddings from Code Generation Models
Paper • 2508.21290 • Published • 21 -
Diffusion is a code repair operator and generator
Paper • 2508.11110 • Published
-
NanoResearch: Co-Evolving Skills, Memory, and Policy for Personalized Research Automation
Paper • 2605.10813 • Published • 16 -
FashionChameleon: Towards Real-Time and Interactive Human-Garment Video Customization
Paper • 2605.15824 • Published • 67 -
Memory-Efficient Looped Transformer: Decoupling Compute from Memory in Looped Language Models
Paper • 2605.07721 • Published • 29 -
MAP: A Map-then-Act Paradigm for Long-Horizon Interactive Agent Reasoning
Paper • 2605.13037 • Published • 8
-
VESPO: Variational Sequence-Level Soft Policy Optimization for Stable Off-Policy LLM Training
Paper • 2602.10693 • Published • 221 -
Flash-KMeans: Fast and Memory-Efficient Exact K-Means
Paper • 2603.09229 • Published • 85 -
DIVE: Scaling Diversity in Agentic Task Synthesis for Generalizable Tool Use
Paper • 2603.11076 • Published • 5 -
LongCat-Flash-Prover: Advancing Native Formal Reasoning via Agentic Tool-Integrated Reinforcement Learning
Paper • 2603.21065 • Published • 78
-
Kimi Linear: An Expressive, Efficient Attention Architecture
Paper • 2510.26692 • Published • 133 -
GLM-5: from Vibe Coding to Agentic Engineering
Paper • 2602.15763 • Published • 198 -
Believe Your Model: Distribution-Guided Confidence Calibration
Paper • 2603.03872 • Published • 40 -
OpenWorldLib: A Unified Codebase and Definition of Advanced World Models
Paper • 2604.04707 • Published • 203
-
Star Attention: Efficient LLM Inference over Long Sequences
Paper • 2411.17116 • Published • 53 -
MSA: Memory Sparse Attention for Efficient End-to-End Memory Model Scaling to 100M Tokens
Paper • 2603.23516 • Published • 53 -
TriAttention: Efficient Long Reasoning with Trigonometric KV Compression
Paper • 2604.04921 • Published • 117 -
Let ViT Speak: Generative Language-Image Pre-training
Paper • 2605.00809 • Published • 33
-
Your Mixture-of-Experts LLM Is Secretly an Embedding Model For Free
Paper • 2410.10814 • Published • 51 -
Make LoRA Great Again: Boosting LoRA with Adaptive Singular Values and Mixture-of-Experts Optimization Alignment
Paper • 2502.16894 • Published • 33 -
Ring-lite: Scalable Reasoning via C3PO-Stabilized Reinforcement Learning for LLMs
Paper • 2506.14731 • Published • 8 -
SlimMoE: Structured Compression of Large MoE Models via Expert Slimming and Distillation
Paper • 2506.18349 • Published • 13