Papers

• mSFT: Addressing Dataset Mixtures Overfiting Heterogeneously in Multi-task SFT (2026)
  Woosung Koh et al.
  1.94
• Building a Precise Video Language with Human-AI Oversight (2026)
  Zhiqiu Lin et al.
  1.94
• Orchard: An Open-Source Agentic Modeling Framework (2026)
  Baolin Peng et al.
  1.94
• EnvFactory: Scaling Tool-Use Agents via Executable Environments Synthesis and Robust RL (2026)
  Minrui Xu et al.
  1.89
• FireRed-OCR Technical Report (2026)
  Hao Wu et al.
  1.78
• Critique Fine-Tuning: Learning to Critique is More Effective than Learning to Imitate (2025)
  Yubo Wang et al.
  1.28
• Video-R1: Reinforcing Video Reasoning in MLLMs (2025)
  Kaituo Feng et al.
  1.28
• OpenThinkIMG: Learning to Think with Images via Visual Tool Reinforcement Learning (2025)
  Zhaochen Su et al.
  1.28
• How Should We Enhance the Safety of Large Reasoning Models: An Empirical Study (2025)
  Zhexin Zhang et al.
  1.28
• VGR: Visual Grounded Reasoning (2025)
  Jiacong Wang et al.
  1.28
• Klear-Reasoner: Advancing Reasoning Capability via Gradient-Preserving Clipping Policy Optimization (2025)
  Zhenpeng Su et al.
  1.28
• Mol-R1: Towards Explicit Long-CoT Reasoning in Molecule Discovery (2025)
  Jiatong Li et al.
  1.28
• Advancing Speech Understanding in Speech-Aware Language Models with GRPO (2025)
  Avishai Elmakies et al.
  1.28
• FrameThinker: Learning to Think with Long Videos via Multi-Turn Frame Spotlighting (2025)
  Zefeng He et al.
  1.28
• Agent Data Protocol: Unifying Datasets for Diverse, Effective Fine-tuning of LLM Agents (2025)
  Yueqi Song et al.
  1.28
• MIRA: Multimodal Iterative Reasoning Agent for Image Editing (2025)
  Ziyun Zeng et al.
  1.28
• InfiniteVL: Synergizing Linear and Sparse Attention for Highly-Efficient, Unlimited-Input Vision-Language Models (2025)
  Hongyuan Tao et al.
  1.28
• Efficient RLHF: Reducing the Memory Usage of PPO (2023)
  Michael Santacroce et al.
  —
• Model Surgery: Modulating LLM's Behavior Via Simple Parameter Editing (2024)
  Huanqian Wang et al.
  —
• Mulberry: Empowering MLLM with o1-like Reasoning and Reflection via Collective Monte Carlo Tree Search (2024)
  Huanjin Yao et al.
  —