SOTAVerified

MME

MME is a comprehensive evaluation benchmark for multimodal large language models. It measures both perception and cognition abilities on a total of 14 subtasks, including existence, count, position, color, poster, celebrity, scene, landmark, artwork, OCR, commonsense reasoning, numerical calculation, text translation, and code reasoning.

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1–25 of 95 papers

TitleStatusHype
High-Resolution Visual Reasoning via Multi-Turn Grounding-Based Reinforcement LearningCode2
Flash-VStream: Efficient Real-Time Understanding for Long Video StreamsCode3
Q-Frame: Query-aware Frame Selection and Multi-Resolution Adaptation for Video-LLMs—0
VideoDeepResearch: Long Video Understanding With Agentic Tool UsingCode2
Language-Vision Planner and Executor for Text-to-Visual Reasoning—0
DynTok: Dynamic Compression of Visual Tokens for Efficient and Effective Video Understanding—0
Fast or Slow? Integrating Fast Intuition and Deliberate Thinking for Enhancing Visual Question Answering—0
SiLVR: A Simple Language-based Video Reasoning FrameworkCode1
EvoMoE: Expert Evolution in Mixture of Experts for Multimodal Large Language Models—0
MME-Reasoning: A Comprehensive Benchmark for Logical Reasoning in MLLMs—0
Enhancing Visual Reliance in Text Generation: A Bayesian Perspective on Mitigating Hallucination in Large Vision-Language Models—0
VideoEval-Pro: Robust and Realistic Long Video Understanding EvaluationCode4
Mitigating Hallucinations via Inter-Layer Consistency Aggregation in Large Vision-Language Models—0
VISTA: Enhancing Vision-Text Alignment in MLLMs via Cross-Modal Mutual Information Maximization—0
Visual Instruction Tuning with Chain of Region-of-Interest—0
TimeChat-Online: 80% Visual Tokens are Naturally Redundant in Streaming VideosCode3
FRAG: Frame Selection Augmented Generation for Long Video and Long Document UnderstandingCode1
An LMM for Efficient Video Understanding via Reinforced Compression of Video Cubes—0
Eagle 2.5: Boosting Long-Context Post-Training for Frontier Vision-Language ModelsCode4
MME-Unify: A Comprehensive Benchmark for Unified Multimodal Understanding and Generation Models—0
SpaceR: Reinforcing MLLMs in Video Spatial ReasoningCode2
BOLT: Boost Large Vision-Language Model Without Training for Long-form Video UnderstandingCode1
Instruction-Aligned Visual Attention for Mitigating Hallucinations in Large Vision-Language ModelsCode0
Improving LLM Video Understanding with 16 Frames Per Second—0
Logic-in-Frames: Dynamic Keyframe Search via Visual Semantic-Logical Verification for Long Video Understanding—0
Show:102550
← PrevPage 1 of 4Next →

No leaderboard results yet.