mlsanityData to 7 Oct 2026

Are We on the Right Way for Evaluating Large Vision-Language Models?

arXiv 2403.20330 ↗First posted 29 Mar 2024cs.CVcited by 893

On the map: Multimodal LLMs / vision-language models → VLM benchmarks and visual reasoning

Abstract (arXiv)

Large vision-language models (LVLMs) have recently achieved rapid progress, sparking numerous studies to evaluate their multi-modal capabilities. However, we dig into current evaluation works and identify two primary issues: 1) Visual content is unnecessary for many samples. The answers can be directly inferred from the questions and options, or the world knowledge embedded in LLMs. This phenomenon is prevalent across current benchmarks. For instance, GeminiPro achieves 42.9% on the MMMU benchmark without any visual input, and outperforms the random choice baseline across six benchmarks over 24% on average. 2) Unintentional data leakage exists in LLM and LVLM training. LLM and LVLM could still answer some visual-necessary questions without visual content, indicating the memorizing of these samples within large-scale training data. For example, Sphinx-X-MoE gets 43.6% on MMMU without accessing images, surpassing its LLM backbone with 17.9%. Both problems lead to misjudgments of actual multi-modal gains and potentially misguide the study of LVLM. To this end, we present MMStar, an elite vision-indispensable multi-modal benchmark comprising 1,500 samples meticulously selected by humans. MMStar benchmarks 6 core capabilities and 18 detailed axes, aiming to evaluate LVLMs' multi-modal capacities with carefully balanced and purified samples. These samples are first roughly selected from current benchmarks with an automated pipeline, human review is then involved to ensure each curated sample exhibits visual dependency, minimal data leakage, and requires advanced multi-modal capabilities. Moreover, two metrics are developed to measure data leakage and actual performance gain in multi-modal training. We evaluate 16 leading LVLMs on MMStar to assess their multi-modal capabilities, and on 7 benchmarks with the proposed metrics to investigate their data leakage and actual multi-modal gain.

Key facts. Cited by 893 papers in the map (data through 7 Oct 2026). Placed in VLM benchmarks and visual reasoning, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 43 of its 67 references that are in the map. First posted on arXiv 29 Mar 2024.

Builds on

43 of 67 references.

Closest ancestors · 1

Same direction, newest first

  1. 2023MathVista: Evaluating Mathematical Reasoning of Foundation Models in Visual Contexts3 Oct

Other references · 42

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2020Language Models are Few-Shot Learnerscited by 32,199
  3. 2022Training language models to follow instructions with human feedbackcited by 15,712
  4. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 12,865
  5. 2023Visual Instruction Tuningcited by 7,792
  6. 2023BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Modelscited by 6,069
  7. 2022PaLM: Scaling Language Modeling with Pathwayscited by 5,304
  8. 2023Improved Baselines with Visual Instruction Tuningcited by 4,362
  9. 2021Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervisioncited by 3,848
  10. 2023Qwen Technical Reportcited by 3,273
  11. 2023Mistral 7Bcited by 3,065
  12. 2023MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Modelscited by 2,689
30 more
  1. 2023InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Taskscited by 2,534
  2. 2016Making the V in VQA Matter: Elevating the Role of Image Understanding in Visual Question Answeringcited by 2,498
  3. 2023MMBench: Is Your Multi-modal Model an All-around Player?cited by 2,105
  4. 2023MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGIcited by 2,044
  5. 2022Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answeringcited by 1,881
  6. 2024Mixtral of Expertscited by 1,785
  7. 2023MME: A Comprehensive Evaluation Benchmark for Multimodal Large Language Modelscited by 1,624
  8. 2023MM-Vet: Evaluating Large Multimodal Models for Integrated Capabilitiescited by 1,221
  9. 2023mPLUG-Owl: Modularization Empowers Large Language Models with Multimodalitycited by 1,038
  10. 2021GLM: General Language Model Pretraining with Autoregressive Blank Infillingcited by 1,026
  11. 2023ShareGPT4V: Improving Large Multi-Modal Models with Better Captionscited by 1,000
  12. 2023SEED-Bench: Benchmarking Multimodal LLMs with Generative Comprehensioncited by 929
  13. 2016A Diagram Is Worth A Dozen Imagescited by 793
  14. 2024DeepSeek-VL: Towards Real-World Vision-Language Understandingcited by 718
  15. 2024Yi: Open Foundation Models by 01.AIcited by 706
  16. 2023CogVLM: Visual Expert for Pretrained Language Modelscited by 704
  17. 2022A-OKVQA: A Benchmark for Visual Question Answering using World Knowledgecited by 702
  18. 2023Baichuan 2: Open Large-scale Language Modelscited by 664
  19. 2024DeepSeek LLM: Scaling Open-Source Language Models with Longtermismcited by 621
  20. 2023OPERA: Alleviating Hallucination in Multi-Modal Large Language Models via Over-Trust Penalty and Retrospection-Allocationcited by 468
  21. 2023Monkey: Image Resolution and Text Label Are Important Things for Large Multi-modal Modelscited by 356
  22. 2024InternLM-XComposer2: Mastering Free-form Text-Image Composition and Comprehension in Vision-Language Large Modelcited by 326
  23. 2023InternLM-XComposer: A Vision-Language Large Model for Advanced Text-image Comprehension and Compositioncited by 297
  24. 2023Q-Bench: A Benchmark for General-Purpose Foundation Models on Low-level Visioncited by 247
  25. 2024Aligning Modalities in Vision Large Language Models via Preference Fine-tuningcited by 197
  26. 2024TinyLLaVA: A Framework of Small-scale Large Multimodal Modelscited by 164
  27. 2023To See is to Believe: Prompting GPT-4V for Better Visual Instruction Tuningcited by 143
  28. 2024SPHINX-X: Scaling Data and Parameters for a Family of Multi-modal Large Language Modelscited by 130
  29. 2023Cheap and Quick: Efficient Vision-Language Instruction Tuning for Large Language Modelscited by 100
  30. 2024Lumen: Unleashing Versatile Vision-Centric Capabilities of Large Multimodal Modelscited by 24

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 139; 2025: 404; 2026: 350.

893 papers.

Sort:

By area:

  1. 2024Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolutioncited by 4,151
  2. 2024LLaVA-OneVision: Easy Visual Task Transfercited by 2,669
  3. 2025Qwen3-VL Technical Reportcited by 2,534
  4. 2025InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Modelscited by 1,769
  5. 2024Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scalingcited by 1,573
  6. 2025InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiencycited by 1,419
  7. 2024Cambrian-1: A Fully Open, Vision-Centric Exploration of Multimodal LLMscited by 822
  8. 2025Qwen2.5-Omni Technical Reportcited by 704
  9. 2025Vision-R1: Incentivizing Reasoning Capability in Multimodal Large Language Modelscited by 653
  10. 2024DeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understandingcited by 559
  11. 2025Qwen3-Omni Technical Reportcited by 451
  12. 2024LLaVA-CoT: Let Vision Language Models Reason Step-by-Stepcited by 446
38 more
  1. 2024VLMEvalKit: An Open-Source Toolkit for Evaluating Large Multi-Modality Modelscited by 443
  2. 2025GLM-4.5V and GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learningcited by 439
  3. 2024LVBench: An Extreme Long Video Understanding Benchmarkcited by 402
  4. 2024ShareGPT4Video: Improving Video Understanding and Generation with Better Captionscited by 386
  5. 2025Kimi-VL Technical Reportcited by 381
  6. 2024LMMs-Eval: Reality Check on the Evaluation of Large Multimodal Modelscited by 344
  7. 2024SpatialRGPT: Grounded Spatial Reasoning in Vision Language Modelscited by 297
  8. 2025R1-VL: Learning to Reason with Multimodal Large Language Models via Step-wise Group Relative Policy Optimizationcited by 296
  9. 2024MME-RealWorld: Could Your Multimodal LLM Challenge High-Resolution Real-World Scenarios that are Difficult for Humans?cited by 284
  10. 2024PyramidDrop: Accelerating Your Large Vision-Language Models via Pyramid Visual Redundancy Reductioncited by 283
  11. 2024We-Math: Does Your Large Multimodal Model Achieve Human-like Mathematical Reasoning?cited by 281
  12. 2025LMM-R1: Empowering 3B LMMs with Strong Reasoning Abilities Through Two-Stage Rule-Based RLcited by 244
  13. 2025Show-o2: Improved Native Unified Multimodal Modelscited by 242
  14. 2025LLaVA-OneVision-1.5: Fully Open Framework for Democratized Multimodal Trainingcited by 233
  15. 2025SmolVLM: Redefining small and efficient multimodal modelscited by 226
  16. 2024CogVLM2: Visual Language Models for Image and Video Understandingcited by 214
  17. 2024InternLM-XComposer-2.5: A Versatile Large Vision Language Model Supporting Long-Contextual Input and Outputcited by 195
  18. 2025Thinking with Images for Multimodal Reasoning: Foundations, Methods, and Future Frontierscited by 193
  19. 2025VITA-1.5: Towards GPT-4o Level Real-Time Vision and Speech Interactioncited by 191
  20. 2025MiniCPM-V 4.5: Cooking Efficient MLLMs via Architecture, Data, and Training Recipecited by 174
  21. 2024VITA: Towards Open-Source Interactive Omni Multimodal LLMcited by 172
  22. 2024Building and better understanding vision-language models: insights and future directionscited by 169
  23. 2024Ovis: Structural Embedding Alignment for Multimodal Large Language Modelcited by 167
  24. 2024OCRBench v2: An Improved Benchmark for Evaluating Large Multimodal Models on Visual Text Localization and Reasoningcited by 165
  25. 2025Thyme: Think Beyond Imagescited by 160
  26. 2024InternLM-XComposer2-4KHD: A Pioneering Large Vision-Language Model Handling Resolutions from 336 Pixels to 4K HDcited by 156
  27. 2025Sa2VA: Marrying SAM2 with MLLM for Dense Grounded Understanding of Images and Videoscited by 154
  28. 2026Qwen3.5-Omni Technical Reportcited by 146
  29. 2025LlamaV-o1: Rethinking Step-by-step Visual Reasoning in LLMscited by 145
  30. 2025LLaDA-V: Large Language Diffusion Models with Visual Instruction Tuningcited by 145
  31. 2024Vision language models are blind: Failing to translate detailed visual features into wordscited by 144
  32. 2025SoTA with Less: MCTS-Guided Sample Selection for Data-Efficient Visual Reasoning Self-Improvementcited by 144
  33. 2024AuroraCap: Efficient, Performant Video Detailed Captioning and a New Benchmarkcited by 137
  34. 2024Improve Vision Language Model Chain-of-thought Reasoningcited by 135
  35. 2025OVO-Bench: How Far is Your Video-LLMs from Real-World Online Video Understanding?cited by 134
  36. 2025RoboBrain: A Unified Brain Model for Robotic Manipulation from Abstract to Concretecited by 134
  37. 2025MMVU: Measuring Expert-Level Multi-Discipline Video Understandingcited by 131
  38. 2024RLAIF-V: Open-Source AI Feedback Leads to Super GPT-4V Trustworthinesscited by 129

50 most cited of 893 All 893 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2023LVLM-eHub: A Comprehensive Evaluation Benchmark for Large Vision-Language Models
  2. 2023ReForm-Eval: Evaluating Large Vision Language Models via Unified Re-Formulation of Task-Oriented Benchmarks
  3. 2024LM4LV: A Frozen Large Language Model for Low-level Vision Tasks
  4. 2023LOVM: Language-Only Vision Model Selection
  5. 2024Are Bigger Encoders Always Better in Vision Large Models?
  6. 2025Centurio: On Drivers of Multilingual Ability of Large Vision-Language Model
  7. 2024How Well Can Vision Language Models See Image Details?
  8. 2026Benchmarking Deflection and Hallucination in Large Vision-Language Models