| 2026-11-04 | 2611.06483 | What Does Multi-Turn Reinforcement Learning Really Improve in LLM Agents? | Kejian Zhu et al. | arXiv | 36 | |
| 2026-11-03 | 2611.04467 | Is Forgetting Unified? Exploring Concept Unlearning in Unified Multimodal Models | Zhongkai Xue et al. | arXiv | 3 |  |
| 2026-10-28 | 2610.38678 | AskStream: When and What to Ask Proactively under Uncertainty in Streaming Video Understanding | Lu Wang et al. | arXiv | 10 | |
| 2026-10-15 | 2610.21143 | CoSTA: Compressed Speech Token Adaptation for Efficient Spoken Dialogue Models | Yingzhuo Deng et al. | arXiv | 12 | |
| 2026-09-10 | 2609.09113 | SAEScientist-Bench: Can AI Agents Conduct Autonomous SAE Interpretability Research? | Yuqiao Tan et al. | arXiv | 20 | |
| 2026-08-10 | 2608.03573 | SFT Conflicts, RL Coexists: A Theoretical and Empirical Analysis of Multi-Task Learning for LLMs | Kejian Zhu et al. | arXiv | 56 | |
| 2026-08-10 | 2608.03571 | Beyond Simply Environment Scaling: Designing Effective Environment Distributions for Multimodal Agent Learning | Kejian Zhu et al. | arXiv | 41 | |
| 2026-08-07 | 2608.06216 | Continual Learning in Transition | Zhiyan Hou et al. | arXiv | 18 | |
| 2026-08-04 | 2608.02499 | SWE-Touch: Benchmarking Coding Agents When Users Touch the Code | Yuqiao Tan et al. | arXiv | 19 | |
| 2026-07-31 | 2607.28624 | PhiZero: A World Model Built Around Physical Language | Shuyao Shang et al. | arXiv | 104 | |
| 2026-07-28 | 2607.24720 | The Physics of Multi-Turn Long-Horizon Planning: From Pre-training to Post-training via Single- and Multi-Teacher On-Policy Agentic Distillation | Tianyi Men et al. | arXiv | 27 | |
| 2026-06-26 | 2606.26027 | Why Multi-Step Tool-Use Reinforcement Learning Collapses and How Supervisory Signals Fix It | Yupu Hao et al. | arXiv | 19 | |
| 2026-06-25 | 2606.22565 | Look Light, Think Heavy: What Multimodal Chain-of-Thought Reasoning Can and Cannot Do | Zhuoran Jin et al. | arXiv | 8 | |
| 2026-05-26 | 2605.23264 | Coloring the Noise: Adversarial Sobolev Alignment for Faithful Image Super Resolution | Hongbo Wang et al. | arXiv | 3 | |
| 2026-05-12 | 2605.06096 | Uncovering Entity Identity Confusion in Multimodal Knowledge Editing | Shu Wu et al. | arXiv | 1 | |
| 2026-05-07 | 2605.00380 | ResRL: Boosting LLM Reasoning via Negative Sample Projection Residual Reinforcement Learning | Zihan Lin et al. | arXiv | 4 | |