VideoMME

The 'VideoMME' dataset/benchmark contains a collection of videos designed to evaluate models on their ability to perform multi-modal reasoning and understanding tasks in a video context.

Papers using VideoMME (30)

Task-Focused Memorization for Multimodal Agents2026Native Active Perception as Reasoning for Omni-Modal Understanding2026Xiaomi Mimo-vl-miloco Technical Report2025MorphoQuant: Modality-Aware Quantization for Omni-modal Large Language Models2026Reflect-R1: Evidence-Driven Reflection for Self-Correction in Long Video Understanding2026VisReflect: Latent Visual Reflection for Fine-Grained Perception in Long Visual Context2026Dynin-Omni: Omnimodal Unified Large Diffusion Language Model2026Where to Focus: Query-Modulated Multimodal Keyframe Selection for Long Video Understanding2026Event-Anchored Frame Selection for Effective Long-Video Understanding2026EvoPrune: Early-Stage Visual Token Pruning for Efficient MLLMs2026Beyond Single-Sample: Reliable Multi-Sample Distillation for Video Understanding2026Question-guided Visual Compression with Memory Feedback for Long-Term Video Understanding2026HiMu: Hierarchical Multimodal Frame Selection for Long Video Question Answering2026LensWalk: Agentic Video Understanding by Planning How You See in Videos2026MACD: Model-Aware Contrastive Decoding via Counterfactual Data2026MSJoE: Jointly Evolving MLLM and Sampler for Efficient Long-Form Video Understanding2026LinMU: Multimodal Understanding Made Linear2026Scaling Vision Language Models for Pharmaceutical Long Form Video Reasoning on Industrial GenAI Platform2026Video Evidence to Reasoning Efficient Video Understanding via Explicit Evidence Grounding2026Think-Clip-Sample: Slow-Fast Frame Selection for Video Understanding2026LiViBench: An Omnimodal Benchmark for Interactive Livestream Video Understanding2026Structured Over Scale: Learning Spatial Reasoning from Educational Video2026VideoAnchor: Reinforcing Subspace-Structured Visual Cues for Coherent Visual-Spatial Reasoning2025Scaling RL To Long Videos2025VSI: Visual Subtitle Integration For Keyframe Selection To Enhance Long Video Understanding2025Lightweight Structured Multimodal Reasoning For Clinical Scene Understanding In Robotics2025Enhancing Temporal Understanding In Video-llms Through Stacked Temporal Attention In Vision Encoders2025Less Is More: Token-efficient Video-qa Via Adaptive Frame-pruning And Semantic Graph Integration2025Q-Frame: Query-aware Frame Selection and Multi-Resolution Adaptation for Video-LLMs2025Video-MME: The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysis2024 · 11 cites