{"ID":2836769,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.19972","arxiv_id":"2511.19972","title":"Boosting Reasoning in Large Multimodal Models via Activation Replay","abstract":"Recently, Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as an effective approach to incentivizing reasoning capability in Large Multimodal Models (LMMs), while the underlying mechanisms behind this post-training paradigm are poorly understood. We begin by exploring how input activations are affected by RLVR through the perspective of logit lens. Our systematic investigations across multiple post-trained LMMs suggest that RLVR shifts low-entropy activations unexpectedly, while high-entropy ones are less affected. We further demonstrate that such phenomena are associated with LMM reasoning by controlled experiments, suggesting a potentially beneficial role of modulating low-entropy activations. To this end, we propose Activation Replay, a novel simple yet effective training-free approach that boosts multimodal reasoning of post-trained LMMs without requiring expensive policy optimization. Our design involves manipulation of visual tokens at test time, replaying low-entropy activations from the input context of base LMMs to regulating the RLVR counterparts. Activation Replay triggers better reasoning across diverse scenarios, including mathematics, o3-like visual agents, and video reasoning. We further show that Activation Replay boosts Pass@K and mitigates narrower reasoning coverage of RLVR. Our design is compared against alternative choices, such as replaying high-entropy activations instead of low-entropy ones, or direct cross-model intervention instead of manipulating input tokens, demonstrating the superiority of our implementation. Code is publicly available at https://github.com/latentcraft/replay.","short_abstract":"Recently, Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as an effective approach to incentivizing reasoning capability in Large Multimodal Models (LMMs), while the underlying mechanisms behind this post-training paradigm are poorly understood. We begin by exploring how input activations are affected...","url_abs":"https://arxiv.org/abs/2511.19972","url_pdf":"https://arxiv.org/pdf/2511.19972v3","authors":"[\"Yun Xing\",\"Xiaobin Hu\",\"Qingdong He\",\"Jiangning Zhang\",\"Shuicheng Yan\",\"Shijian Lu\",\"Yu-Gang Jiang\"]","published":"2025-11-25T06:31:57Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"Reinforcement Learning\"]","has_code":false,"code_links":[{"ID":606627,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2836769,"paper_url":"https://arxiv.org/abs/2511.19972","paper_title":"Boosting Reasoning in Large Multimodal Models via Activation Replay","repo_url":"https://github.com/latentcraft/replay","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}