Correcting What You Cannot See: Credit Assignment for Perception Distillation in Multimodal Reasoners

Excerpt

arXiv:2607.28336v3 Announce Type: replace Abstract: On-policy distillation provides dense supervision for multimodal reasoners, but its trajectory-level reward cannot determine whether a failed answer arose from perception or subsequent reasoning. Perception Success Rate (PSR), estimated from multiple reasonings sharing one perception, remains ambiguous because low success conflates perceptual insufficiency with reasoning difficulty. We introduce \textbf{Perception-Correction Distillation (PCD)}