Safety
Video-OPSD: Exploiting Privileged Visual Evidence for On-Policy Self-Distillation in Video Large Language Models
arXiv:2608.27065v1 Announce Type: new Abstract: On-policy self-distillation (OPSD) has recently emerged as an effective post-training paradigm that improves policy optimization through dense token-lev
arXiv:2608.27065v1 Announce Type: new Abstract: On-policy self-distillation (OPSD) has recently emerged as an effective post-training paradigm that improves policy optimization through dense token-level supervision from a privileged self-teacher. Despite its promise, OPSD remains largely underexplored for Video Large Language Models (Video-LLMs). Existing methods typically construct privileged teachers by augmenting their context with additional information while keeping the primary input unchanged for both teacher and student. Video reasoning, however, offers a distinct source of privileged supervision within the primary input itself: long videos contain substantial temporal redundancy, and only a small subset of frames provides the evidence necessary to answer a question. Building on this observation, we present extbf{Video-OPSD}, an OPSD framework that exploits privileged visual evidence for both self-teacher construction and knowledge transfer. First, our Evidence-Grounded Self-Teacher conditions the teacher exclusively on annotated evidence frames while the student continues to reason over the complete video. This focused visual input enables the teacher to provide more informative supervision. Second, our Evidence-Guided Token Optimization adaptively weights token-level distillation according to each reasoning token's reliance on privileged visual evidence, thereby emphasizing perceptually grounded reasoning. Experiments across video understanding and reasoning benchmarks show that extbf{Video-OPSD} consistently improves upon Standard OPSD across multiple backbones and achieves performance comparable to GRPO while requiring substantially less training time, establishing an effective and efficient post-training approach for Video-LLMs.
Related
- OPSD-V: On-Policy Self-Distillation for Post-Training Few-Step Autoregressive Video Generators
- Video-OPD: Efficient Post-Training of Multimodal Large Language Models for Temporal Video Grounding via On-Policy Distillation
- Visual-OPSD: Cross-Modal On-Policy Self-Distillation for Efficient Unified Multimodal Reasoning
Source: arXiv cs.CV | 2026-08-28