CORTEXA
← Browse
arxivcs.CV2026-07-06

Claim-Level Rubric Rewards for Video Caption Reinforcement Learning

Mingqi Gao, Hongyuan Dong, Yifei Chen, Zhisheng Zhong, Zheng Ruan, Wenjin Hou, Yu Chen, Han Hu, Yansong Tang

In this paper, we introduce Claim-Level Rubric Rewards (CuRe), a structured reward framework designed to address the reward-design bottleneck in reinforcement learning for dense video captioning. Existing reward designs generally fall into two categories: holistic response-level judgment across heterogeneous criteria, or alignment-based evaluation against reference captions. However, both paradigms suffer from fundamental limitations. Holistic rewards struggle to ensure factual accuracy and are prone to stylistic reward hacking, while reference-based rewards overly rely on rigid textual alignment, failing to preserve the completeness and diversity inherent to open-ended generation tasks. To address these challenges, CuRe reformulates reward modeling as fine-grained claim-level verification. Specifically, CuRe decomposes captions into category-aware atomic claims through a structured rubric, converting holistic evaluation into simpler and more reliable claim-level verification.

View free PDFSource page

Related papers

arxivcs.LGcs.CV2026-06-26

NormGuard: Reward-Preserving Norm Constraints in Flow-Matching Reinforcement Learning

Tianlin Pan, Lianyu Pang, Cheng Da, Huan Yang, Changqian Yu, Kun Gai, et al.

Reinforcement learning (RL) post-training improves the reward alignment of flow-based generators, but often degrades perceptual quality in ways that are not captured by the reward proxy. We identify a simple structural signature of this drift: across three post-training methods (…

View free PDFSource page
arxivcs.CV2026-07-22

PercepCap: Video Captioner with Structured Spatio-Temporal Perception

Yifan Xu, Zihao Wang, Zhixiao Wang, Jiaming Zhang, Yichun Yang, Desen Meng, et al.

Video captioning requires fine-grained spatio-temporal understanding of videos, including spatial perception of where objects are located and temporal perception of when events occur. Existing MLLMs usually generate captions directly from video inputs without exposing the percept…

View free PDFSource page
arxivcs.CV2026-07-15

SIVA-RL: Sensitivity-Invariance Visual Alignment for Multimodal Reinforcement Learning

Cheng Tang, Junzhi Ning, Min Cen, Wei Li, Xinyi Zeng, Pinxian Zeng, et al.

Reinforcement learning with verifiable rewards (RLVR) drives multimodal reasoning, but answer-level correctness does not guarantee that a vision-language model grounds its predictions in visual evidence. Existing visual-intervention methods contrast policy behavior on original an…

View free PDFSource page
arxivcs.CV2026-07-16

Reasoning-Guided Part-Level Visual Grounding via Reinforcement Learning

Kazi Sajeed Mehrab, Hani Alomari, Najibul Haque Sarker, Chia-Wei Tang, Zaber Ibn Abdul Hakim, Anuj Karpatne, et al.

Multimodal large language models (MLLMs) ground whole objects well from free-form language queries, but they struggle when the query names a part rather than the object. We trace this to a missing object-part hierarchy, since parts are localized in the same single step used for o…

View free PDFSource page
arxivcs.CV2026-07-14

AVSCap: Orchestrating Audio-Visual Synergy for Omni-modal Video Captioning

Yanghai Wang, Jiahao Wang, Jiafu Tang, Yuanxing Zhang, Zhe Cao, Hanyan Bian, et al.

Omni-modal video captioning is not merely combining visual captioning with audio transcription: a useful caption must describe how visual actions, speech, music, and sound effects co-evolve. Existing large multimodal models often fail at this relational step, treating audio and v…

View free PDFSource page
arxivcs.CVcs.AIcs.MM2026-07-03

Parallelized Autoregressive Decoding for Omni-Modal Dense Video Captioning

Wenzheng Zeng, Siyi Jiao, Chen Gao, Hwee Tou Ng, Mike Zheng Shou

Dense video captioning aims to generate temporally grounded descriptions of video events, benefiting both event-level video understanding and generation. In this domain, autoregressive video large language models have emerged as a prevalent paradigm due to their strong generative…

View free PDFSource page