Annotations as Rollouts: Efficient and Scalable Reinforcement Learning for Video MLLMs
- 类型：arxiv
- 标识：2608.20492
- 链接：https://arxiv.org/abs/2608.20492
- 主分类：multimodal
- 形态：method
- TLDR：Multimodal large language models (MLLMs) have become a prevailing paradigm for unified video perception. However, post-training on large multi-task datasets remains challenging, as existing reinforcement learning methods sample on-policy groups with few high-quality rollouts even with costly chain-of-thought (CoT) generation. In this paper, we study the sample efficiency and scalability of RL post-training for video MLLMs and introduce OraRL. We identify an overlooked role for annotations: Beyond scoring rollouts, each can enter its on-policy group as an oracle rollout, a direct positive optim
- 副分类：engineering
- 待LLM分类：否
- 来源文件：
- /inbox/tom/_candidates/2026-08-26-agent-rag-longcontext-candidates.json