Watch-Think-Interact: Bootstrapping Long-Horizon Multi-Turn Streaming Video Reasoning with Reinforcement Learning
cs.AI
Submitted: 2026-09-29
Updated: 2026-09-29
Terminology
Sources
- Qwen3-VL Technical Report
- Qwen2.5-VL Technical Report
- MiniCPM-o 4.5: Towards Real-Time Full-Duplex Omni-Modal Interaction
- GPT-4o System Card
- StreamingBench: Assessing the Gap for MLLMs to Achieve Streaming Video Understanding
- GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimization
- LongVU: Spatiotemporal Adaptive Compression for Long Video-Language Understanding
- Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context
- Streaming Video Instruction Tuning
- Group Sequence Policy Optimization
- VideoSearcher: Empowering Video Deep Research with Multi-Tool Agentic Reasoning via Reinforcement Learning
- TPRU: Advancing Temporal and Procedural Understanding in Large Multimodal Models
- Explore with Long-term Memory: A Benchmark and Multimodal LLM-based Reinforcement Learning Framework for Embodied Exploration
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems