MetaQuery Instruction Tuning Data Collection We downsample high-resolution images so that the shorter side is 1024 pixels (MetaQuery_Instruct_2.4M) or 512 pixels (MetaQuery_Instruct_2.4M_512res) • 2 items • Updated Jun 24, 2025 • 2
Memory is Reconstructed, Not Retrieved: Graph Memory for LLM Agents Paper • 2606.06036 • Published 18 days ago • 71
Learning from the Self-future: On-policy Self-distillation for dLLMs Paper • 2606.18195 • Published 6 days ago • 72
FORT-Searcher: Synthesizing Shortcut-Resistant Search Tasks for Training Deep Search Agents Paper • 2606.12087 • Published 12 days ago • 75
VibeThinker-3B: Exploring the Frontier of Verifiable Reasoning in Small Language Models Paper • 2606.16140 • Published 7 days ago • 102
DreamX-World 1.0: A General-Purpose Interactive World Model Paper • 2606.16993 • Published 7 days ago • 106
OmniDirector: General Multi-Shot Camera Cloning without Cross-Paired Data Paper • 2606.13432 • Published 11 days ago • 105
LoopCoder-v2: Only Loop Once for Efficient Test-Time Computation Scaling Paper • 2606.18023 • Published 6 days ago • 197
Embodied-R1.5: Evolving Physical Intelligence via Embodied Foundation Models Paper • 2606.11324 • Published 13 days ago • 165
JoyAI-VL-Interaction: Real-Time Vision-Language Interaction Intelligence Paper • 2606.14777 • Published 12 days ago • 195
Fast-WAM: Do World Action Models Need Test-time Future Imagination? Paper • 2603.16666 • Published Mar 17 • 2
RODS: Reward-Driven Online Data Synthesis for Multi-Turn Tool-Use Agents Paper • 2606.19047 • Published 5 days ago • 4
Native Active Perception as Reasoning for Omni-Modal Understanding Paper • 2606.19341 • Published 5 days ago • 14
MolmoMotion: Forecasting Point Trajectories in 3D with Language Instruction Paper • 2606.18558 • Published 5 days ago • 46
Laguna M.1 Collection Our most capable model to date, designed for long-horizon work. Apache 2.0. • 4 items • Updated about 23 hours ago • 15