MMMU

A massive multi-discipline multimodal benchmark of college-level questions requiring image understanding plus expert reasoning.

Papers using MMMU (34)

STEP3-VL-10B Technical Report2026Qwen3-VL Technical Report2025 · 12 citesMorphoQuant: Modality-Aware Quantization for Omni-modal Large Language Models2026Internvl3.5: Advancing Open-source Multimodal Models In Versatility, Reasoning, And Efficiency2025 · 2 citesChatVLA: Unified Multimodal Understanding and Robot Control with Vision-Language-Action Model2025 · 2 citesData-Driven Calibration of Prediction Sets in Large Vision-Language Models Based on Inductive Conformal Prediction2025 · 1 citesTVI-CoT: Text-Visual Interleaved Chain-of-Thought Reasoning for Multimodal Understanding2026Vision Verification Enhanced Fusion of VLMs for Efficient Visual Reasoning2026Same Answer, Different Representations: Hidden instability in VLMs2026LinMU: Multimodal Understanding Made Linear2026Thinking with Video: Video Generation as a Promising Multimodal Reasoning Paradigm2025VisPlay: Self-Evolving Vision-Language Models from Images2025Simple Vision-language Math Reasoning Via Rendered Text2025Diagnosing Visual Reasoning: Challenges, Insights, and a Path Forward2025Fusion to Enhance: Fusion Visual Encoder to Enhance Multimodal Language Model2025LLaVA-Critic-R1: Your Critic Model is Secretly a Strong Policy Model2025SAIL-VL2 Technical Report2025Training Vision-Language Process Reward Models for Test-Time Scaling in Multimodal Reasoning: Key Insights and Lessons Learned2025CoRGI: Verified Chain-of-Thought Reasoning with Post-hoc Visual Grounding2025Multimodal Chain of Continuous Thought for Latent-Space Reasoning in Vision-Language Models2025Reinforcement Fine-Tuning Naturally Mitigates Forgetting in Continual Post-Training2025Skywork-r1v3 Technical Report2025ASCD: Attention-Steerable Contrastive Decoding for Reducing Hallucination in MLLM2025Constructive Distortion: Improving Mllms With Attention-guided Image Warping2025How Far Have Medical Vision-language Models Come? A Comprehensive Benchmarking Study2025Mmjee-eval: A Bilingual Multimodal Benchmark For Evaluating Scientific Reasoning In Vision-language Models2025Test-time Warmup For Multimodal Large Language Models2025Gam-agent: Game-theoretic And Uncertainty-aware Collaboration For Complex Visual Reasoning2025LaViDa: A Large Diffusion Language Model for Multimodal Understanding2025Skywork R1V: Pioneering Multimodal Reasoning with Chain-of-Thought2025MOVE: A Mixture-of-Vision-Encoders Approach for Domain-Focused Vision-Language Processing2025Are We on the Right Way for Evaluating Large Vision-Language Models?2024 · 44 citesMMMU-Pro: A More Robust Multi-discipline Multimodal Understanding Benchmark2024 · 3 citesText as Images: Can Multimodal Large Language Models Follow Printed Instructions in Pixels?2023