Turns Matter: Advancing Streaming Video RL via Turn-Level Credit
cs.CV, cs.CL, cs.LG
Submitted: 2026-11-04
Updated: 2026-11-04
Terminology
Sources
- Qwen3-VL Technical Report
- Qwen2.5-VL Technical Report
- RL Starts before RL: On Policy Distillation for Better Reinforcement Learning
- Beyond Visual Memory: Mechanistic Diagnostics of Latent Visual Reasoning
- StreamPro: From Reactive Perception to Proactive Decision-Making in Streaming Video
- VideoChat-R1: Enhancing Spatio-Temporal Perception via Reinforcement Fine-Tuning
- Thinking in Streaming Video
- gpt-oss-120b & gpt-oss-20b Model Card
- Proximal Policy Optimization Algorithms
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- ProactiveVideoQA: A Comprehensive Benchmark Evaluating Proactive Interactions in Video Large Language Models
- DOPD: Dual On-policy Distillation
- Tarsier2: Advancing Large Vision-Language Models from Detailed Video Description to Comprehensive Video Understanding
- OneStreamer: Unifying Perception, Memory, and Proactive Response in Streaming Video Interaction
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs