method · first seen 19 Oct 2026
reinforcement learning from reward models for video captioning
Also: RL fine-tuning for captioning
Optimizing multimodal captioning models with reinforcement learning using task-specific rewards instead of checklists.
Mentions per week
PapersCodeDiscussion
Earliest sources
- AVTime: Reinforcing Dense Audio-Visual Captions with Bidirectional AlignmentPapers · 19 Oct 2026 · Mingyang Wu, Kaituo Feng, Bohao Li, Kaixiong Gong, Zihao Yin, Xiangyu Yue
Latest sources
- AVTime: Reinforcing Dense Audio-Visual Captions with Bidirectional AlignmentPapers · 19 Oct 2026 · Mingyang Wu, Kaituo Feng, Bohao Li, Kaixiong Gong, Zihao Yin, Xiangyu Yue