The paper introduces CoFire, a reinforcement learning framework for video understanding that addresses the problem of confidence erosion—where initial predictions lose reliability during iterative refinement. CoFire enforces a "confident first, reflective second" principle, ensuring the initial answer remains trustworthy and that subsequent reflection only improves upon it. The method achieves state‑of‑the‑art accuracy on Video‑MME, MLVU, and LongVideoBench, outperforming prior 7B models by 1.7 points and improving both initial and final accuracy over outcome‑reward RL by 4.1 and 3.7 points respectively.
By Zihao Lin, Jingyuan Liu, Yucheng Zhang, Yuguang Yao, Lifu Huang
arXiv:2606. 27922v1 Announce Type: cross Abstract: Current multimodal reflection mechanisms for long video understanding predominantly rely on closed-loop self-reflection within internal parameters.
By Shuimu Chen, Yuteng Chen, Yuanshen Guan, Zebang Cheng, Zeyu Zhang, Shengqian Qin, Bin Xia, Jiaran Li, Wenming Yang, Fei Ma
arXiv:2610.01973v1 Announce Type: new
Abstract: Reinforcement learning (RL) for video generation usually assigns one scalar reward to an entire sampled video. Yet a video is not uniformly flawed: som...
By Yifan Wang, Gordon Guocheng Qian, Yanyu Li, Anil Kag, Yun Fu
The paper introduces RARM, a Reference‑Anchored Reward Model that uses a single successful demonstration to generate dense, progress‑aware rewards for reinforcement learning in robot manipulation. RARM is trained on general‑purpose videos with a contrastive temporal objective, requiring no task‑specific data or reward engineering. During deployment it matches rollout clips to reference clips and rewards only confident forward progress, reducing false positives. Experiments on nine simulated tasks and four real‑world tasks show that RARM achieves the best overall success rates, especially on long‑horizon tasks like cloth folding.
By Pengzhi Yang, Xinyu Wang, Pengyu Jing, Kehan Wen, Yiduo Qu, Zhenhao Huang, Minghao Fu, Xin Liu, Yaheng Shen, Fan Shi
arXiv:2608. 04964v1 Announce Type: new Abstract: Interactive video world models are essential for long-horizon planning and exploration, yet they suffer from compounding errors.
By Bohai Gu, Yueyang Yuan, Taiyi Wu, Dazhao Du, Jian Liu, Xiaoyi Pang, Jie Zhang, Xiaocheng Lu, Haobin Zhong, Xiaotong Zhao, Alan Zhao, Song Guo
arXiv:2606. 29892v1 Announce Type: cross Abstract: Reinforcement learning (RL) has become indispensable for pushing Vision-Language-Action Models (VLAs) beyond static imitation learning.
By Siyao Chen, Jiakang Yuan, Jiaxin Wang, Tao Chen
arXiv:2610.32222v1 Announce Type: new
Abstract: Reinforcement learning has gained attention for improving action-conditioned world models (ACWMs), as it enables reward-driven optimization of visual q...
By Shibo Hong, Siyu Ren, FengJiao Chen, Zhongyuan Peng, Wei Shi, Yixin Cao
arXiv:2610.31820v1 Announce Type: new
Abstract: Reliable confidence estimation is essential for trustworthy deployment of large language models (LLMs), but dedicated confidence training stage require...
By Shujin Wu, Heng Ji
arXiv:2602. 13602v2 Announce Type: replace-cross Abstract: We present \revise (\underline{Re}asoning with \underline{Vi}deo \underline{S}parsity), a multi-round agent for video question answering (VQA).
By Chenwei Xu, Zhen Ye, Shang Wu, Weijian Li, Zihan Wang, Zhuofan Xia, Lie Lu, Pranav Maneriker, Fan Du, Manling Li, Han Liu
SCARL (Sequence Confidence-Aligned Reinforcement Learning) addresses a misalignment between a reinforcement learning policy’s confidence and its reward signal in language-model reasoning. By reshaping each advantage with an alignment gap that reflects the difference between a response’s group-relative confidence and the outcome it certifies, SCARL drives responses with equal advantage toward equal confidence and mitigates entropy collapse. Experiments on Qwen3‑4B‑Base show a 2.4‑point lift in Pass@32 over DAPO and consistent improvements across multiple estimators and backbones while preserving policy entropy.
By Zhenghong Ke, Jingbo Zhou, Hua Wu
arXiv:2610.18614v1 Announce Type: new
Abstract: Controlling an acting LLM agent means deciding, at each step of a long-horizon task, whether to trust its next action, abstain, or hand off to a human....
By Ashutosh Hathidara
arXiv:2611.06728v1 Announce Type: cross
Abstract: Streaming video understanding requires models to repeatedly decide whether to respond or continue observing. Recently, Group Relative Policy Optimiza...
By Shuai Dong, Kanghui Tian, Linghan Chen, Wenqi Shao, Hongrui Wu, Haowen Hou, Yongfu Zhu, Xuehui Wang, Ziyue Wang, Wenpu Liu, Weichu Xie, Yuqi Xu, Congcong Wang, Kaiwen Tuo, Caoyuan Ma, Taiqiang Wu, Ji Zhao, Wenzheng Chang, Dianyi Wang, Siyuan Wang, Zhongyu Wei, Yu Qiao, Jiaqi Wang