V-STORM: A Structure-Aware Grounded Chain of Thought Dataset for Faithful and Fine-Grained Spatio-Temporal Video Reasoning
cs.CV, cs.AI
Submitted: 2026-11-06
Updated: 2026-11-06
Terminology
Sources
- Qwen3-VL Technical Report
- Qwen2.5-VL Technical Report
- Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling
- Video-R1: Reinforcing Video Reasoning in MLLMs
- Chain-of-Frames: Advancing Video Understanding in Multimodal LLMs via Frame-Aware Reasoning
- GPT-4o System Card
- Open-o3-Video: Grounded Video Reasoning with Explicit Spatio-Temporal Evidence
- VCR-Bench: A Comprehensive Evaluation Framework for Video Chain-of-Thought Reasoning
- SAM 2: Segment Anything in Images and Videos
- Gemini: A Family of Highly Capable Multimodal Models
- Traceable Evidence Enhanced Visual Grounded Reasoning: Evaluation and Methodology
- Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution
- VideoRFT: Incentivizing Video Reasoning Capability in MLLMs via Reinforced Fine-Tuning
- InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency
- InternVid: A Large-scale Video-Text Dataset for Multimodal Understanding and Generation
- EOC-Bench: Can MLLMs Identify, Recall, and Forecast Objects in an Egocentric World?
- VideoLLaMA 3: Frontier Multimodal Foundation Models for Image and Video Understanding
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