ReactionBench: A Human-Anchored Benchmark for Modeling Reactions to Video
cs.CV, cs.AI
Submitted: 2026-10-11
Updated: 2026-10-11
Code: https://github.com/MIT-MI/ReactionBench
Project page: https://yibozh.github.io/ReactionBench-page
Terminology
Sources
- Seamless Interaction: Dyadic Audiovisual Motion Modeling and Large-Scale Dataset
- The Alternative Annotator Test for LLM-as-a-Judge: How to Statistically Justify Replacing Human Annotators with LLMs
- AHA -- Predicting What Matters Next: Online Highlight Detection Without Looking Ahead
- MLLM-as-a-Judge: Assessing Multimodal LLM-as-a-Judge with Vision-Language Benchmark
- CG-Bench: Clue-grounded Question Answering Benchmark for Long Video Understanding
- MultiEmo-Bench: Multi-label Visual Emotion Analysis for Multi-modal Large Language Models
- Lost in Time: A New Temporal Benchmark for VideoLLMs
- DeepSeek-V3 Technical Report
- Video-MME-v2: Towards the Next Stage in Benchmarks for Comprehensive Video Understanding
- Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
- A Survey on LLM-as-a-Judge
- StimuVAR: Spatiotemporal Stimuli-aware Video Affective Reasoning with Multimodal Large Language Models
- LongInsightBench: A Comprehensive Benchmark for Evaluating Omni-Modal Models on Human-Centric Long-Video Understanding
- WorldSense: Evaluating Real-world Omnimodal Understanding for Multimodal LLMs
- MotionBench: Benchmarking and Improving Fine-grained Video Motion Understanding for Vision Language Models
- LoRA: Low-Rank Adaptation of Large Language Models
- EmoBench-M: Benchmarking Emotional Intelligence for Multimodal Large Language Models
- EmoTrans: A Benchmark for Understanding, Reasoning, and Predicting Emotion Transitions in Multimodal LLMs
- Video-MMMU: Evaluating Knowledge Acquisition from Multi-Discipline Professional Videos
- LLMs Capture Emotion Labels, Not Emotion Uncertainty: Distributional Analysis and Calibration of Human-LLM Judgment Gaps
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