# MMGist: A Comprehensive Multimodal Benchmark for 2027

arXiv: https://arxiv.org/abs/2606.22437 · HTML: https://mlsanity.com/paper/2606.22437/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 38 references that are in the map.
- First posted on arXiv 21 Jun 2026.

## Builds on (most cited first, 19 of 19)

- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGI](https://mlsanity.com/paper/2311.16502/) (2023)
- [MathVista: Evaluating Mathematical Reasoning of Foundation Models in Visual Contexts](https://mlsanity.com/paper/2310.02255/) (2023)
- [InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency](https://mlsanity.com/paper/2508.18265/) (2025)
- [A Diagram Is Worth A Dozen Images](https://mlsanity.com/paper/1603.07396/) (2016)
- [Measuring Multimodal Mathematical Reasoning with MATH-Vision Dataset](https://mlsanity.com/paper/2402.14804/) (2024)
- [MMMU-Pro: A More Robust Multi-discipline Multimodal Understanding Benchmark](https://mlsanity.com/paper/2409.02813/) (2024)
- [SLAKE: A Semantically-Labeled Knowledge-Enhanced Dataset for Medical Visual Question Answering](https://mlsanity.com/paper/2102.09542/) (2021)
- [Pervasive Label Errors in Test Sets Destabilize Machine Learning Benchmarks](https://mlsanity.com/paper/2103.14749/) (2021)
- [tinyBenchmarks: evaluating LLMs with fewer examples](https://mlsanity.com/paper/2402.14992/) (2024)
- [LogicVista: Multimodal LLM Logical Reasoning Benchmark in Visual Contexts](https://mlsanity.com/paper/2407.04973/) (2024)
- [MedXpertQA: Benchmarking Expert-Level Medical Reasoning and Understanding](https://mlsanity.com/paper/2501.18362/) (2025)
- [Teaching CLIP to Count to Ten](https://mlsanity.com/paper/2302.12066/) (2023)
- [Investigating Data Contamination in Modern Benchmarks for Large Language Models](https://mlsanity.com/paper/2311.09783/) (2023)
- [BetterBench: Assessing AI Benchmarks, Uncovering Issues, and Establishing Best Practices](https://mlsanity.com/paper/2411.12990/) (2024)
- [DyVal: Dynamic Evaluation of Large Language Models for Reasoning Tasks](https://mlsanity.com/paper/2309.17167/) (2023)
- [When AI Benchmarks Plateau: A Systematic Study of Benchmark Saturation](https://mlsanity.com/paper/2602.16763/) (2026)
- [Benchmark Designers Should "Train on the Test Set" to Expose Exploitable Non-Visual Shortcuts](https://mlsanity.com/paper/2511.04655/) (2025)

## Built on by (most cited first, 3 of 3)

- [Contrastive On-Policy Distillation](https://mlsanity.com/paper/2607.19046/) (2026)
- [One Perturbation Is Not Enough: Identifiability and Blind Baselines for Behavioral AI Evaluation](https://mlsanity.com/paper/2609.06190/) (2026)
- [What Looks Like a Capability Limit in Vision-Language Models Is a Readout Limit](https://mlsanity.com/paper/2609.27408/) (2026)

## Abstract (arXiv)

We conduct a systematic study of 18 widely used vision-language benchmarks and identify three major issues: 1) many items do not rely on visual cues and therefore fail to effectively measure multimodal understanding; 2) many items are already close to performance saturation for current LVLMs, which limits their discriminative power; 3) a small number of anomalous items affect the reliability of evaluation results. To this end, we propose MMGist, a curated benchmark that covers seven capability dimensions and contains 7,262 items. MMGist is constructed through a three-stage pipeline, which sequentially combines text-ablation filtering, cross-model saturation filtering, and anomaly detection filtering. We conduct extensive experiments on 27 leading LVLMs and compare MMGist with the raw pool of 23,250 items. The results show that MMGist preserves model rankings with high fidelity, with Spearman ρ = 0.98, while reducing evaluation items by 69% and improving cross-model discrimination by 78%. Further results indicate that Visual Logic remains a systematic weakness of current LVLMs, while knowledge-intensive dimensions such as Expert Knowledge dimensions remain important factors for distinguishing closed-source models from open-source models. These findings suggest that high-quality evaluation should prioritize visual dependency, discriminative power, and reliability, rather than simply pursuing benchmark scale.