Reviso: Not All Visual Evidence Deserves Equal Attention in Video Captioning
cs.CV
Submitted: 2026-11-02
Updated: 2026-11-02
Code: https://github.com/anthropics/claude-code
Terminology
Sources
- LLaVA-OneVision-2: Towards Next-Generation Perceptual Intelligence
- Qwen3-VL Technical Report
- PIGEONS: Incremental Long-Video Captioning with Persistent Identity Gallery and Evolving Overarching Narrative Summary
- Claim-Level Rubric Rewards for Video Caption Reinforcement Learning
- ARC-Hunyuan-Video-7B: Structured Video Comprehension of Real-World Shorts
- RefCaptioner: Multi-Reference Image-Grounded Video Captioning
- LongAV-Compass: Towards Unified Evaluation of Minute-Scale Audio-Visual Generation Across T2AV, I2AV, and V2AV
- Thinking Beyond Videos: Unifying Video Reasoning and Deep Research for Open-World Video Agents
- OmniFysics-Captioner Technical Report: Grounding Omni-Modal Understanding in the Physical World for Better Captioning
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- OmniPack: Unified Token Compression for Efficient Omni-modal Large Language Models
- KeyFrame-Compass: Towards Comprehensive Evaluation of Keyframe-Conditioned Video Generation
- Artifact-Bench: Evaluating MLLMs on Detecting and Assessing the Artifacts of AI-Generated Videos
- Tarsier: Recipes for Training and Evaluating Large Video Description Models
- Beacon: Knowing When and How to Perform Agentic Visual Reasoning
- InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency
- LongCaptioning: Unlocking the Power of Long Video Caption Generation in Large Multimodal Models
- PercepCap: Video Captioner with Structured Spatio-Temporal Perception
- CapRL++: Unified Reinforcement Learning with Verifiable Rewards for Dense Image and Video Captioning
- TimeChat-Captioner: Scripting Multi-Scene Videos with Time-Aware and Structural Audio-Visual Captions
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