pepo-token-level-multimodal-policy

【设计与创意】Replace uniform token-level advantages with perception-exploration gating that weights tokens by visual grounding strength. Adds 3.67 points to geometry reasoning and 5.32 to few-shot classification with <1% compute overhead. Works best for multimodal CoT where visual grounding anchors reasoning. Trigger: When using token-level RL on VLMs and want to…

独立 GitHub 快照 · 自动分类 · 以原作者文档为准