pepo-token-level-multimodal-policy

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 emphas…

Independent GitHub snapshot · Automated classification · Original documentation governs