Beyond Entropy: Self-Diagnostic Multi-Role Token Optimization for Video Reasoning
cs.CV
Submitted: 2026-10-02
Updated: 2026-10-15
Comments: 19 pages, 6 figures, under review
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- Qwen2.5-VL Technical Report
- Video-Holmes: Can MLLM Think Like Holmes for Complex Video Reasoning?
- VideoLLaMA 2: Advancing Spatial-Temporal Modeling and Audio Understanding in Video-LLMs
- The Entropy Mechanism of Reinforcement Learning for Reasoning Language Models
- Reinforcing Video Reasoning with Focused Thinking
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- APPO: Attention-guided Perception Policy Optimization for Video Reasoning
- Video-MMMU: Evaluating Knowledge Acquisition from Multi-Discipline Professional Videos
- GPT-4o System Card
- Entropy Is Not Enough: Unlocking Effective Reinforcement Learning for Visual Reasoning via Vision-Anchored Token Selection
- VideoChat-R1: Enhancing Spatio-Temporal Perception via Reinforcement Fine-Tuning
- Visual-Advantage On-Policy Distillation for Vision-Language Models
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- Video-RTS: Rethinking Reinforcement Learning and Test-Time Scaling for Efficient and Enhanced Video Reasoning
- Video-KTR: Reinforcing Video Reasoning via Key Token Attribution
- Boosting Reasoning in Large Multimodal Models via Activation Replay
- mPLUG-Owl3: Towards Long Image-Sequence Understanding in Multi-Modal Large Language Models
- Not All Tokens See Equally: Perception-Grounded Policy Optimization for Large Vision-Language Models
- Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models