# MLLM-CompBench: A Comparative Reasoning Benchmark for Multimodal LLMs

arXiv: https://arxiv.org/abs/2407.16837 · HTML: https://mlsanity.com/paper/2407.16837/

## Key facts

- Cited by 32 papers in the map (data through 14 Oct 2026).
- Placed in Multimodal reasoning, one of 17 directions in LLM reasoning.
- Builds on 12 of its 70 references that are in the map.
- First posted on arXiv 23 Jul 2024.

## Placed in

- LLM reasoning / [Multimodal reasoning](https://mlsanity.com/lr/multimodal-reasoning/)
- Multimodal LLMs / vision-language models / [VLM benchmarks and visual reasoning](https://mlsanity.com/mllm/vlm-benchmarks-and-visual-reasoning/)

## Builds on (most cited first, 12 of 12)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Hierarchical Text-Conditional Image Generation with CLIP Latents](https://mlsanity.com/paper/2204.06125/) (2022)
- [Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context](https://mlsanity.com/paper/2403.05530/) (2024)
- [Microsoft COCO Captions: Data Collection and Evaluation Server](https://mlsanity.com/paper/1504.00325/) (2015)
- [PaLM 2 Technical Report](https://mlsanity.com/paper/2305.10403/) (2023)
- [Kosmos-2: Grounding Multimodal Large Language Models to the World](https://mlsanity.com/paper/2306.14824/) (2023)
- [Gemini: A Family of Highly Capable Multimodal Models](https://mlsanity.com/paper/2312.11805/) (2023)
- [CogVLM: Visual Expert for Pretrained Language Models](https://mlsanity.com/paper/2311.03079/) (2023)
- [MANTIS: Interleaved Multi-Image Instruction Tuning](https://mlsanity.com/paper/2405.01483/) (2024)
- [ReMI: A Dataset for Reasoning with Multiple Images](https://mlsanity.com/paper/2406.09175/) (2024)
- [Q-Bench+: A Benchmark for Multi-modal Foundation Models on Low-level Vision from Single Images to Pairs](https://mlsanity.com/paper/2402.07116/) (2024)

## Built on by (most cited first, 30 of 32)

- [A Survey on Evaluation of Multimodal Large Language Models](https://mlsanity.com/paper/2408.15769/) (2024)
- [Reinforced MLLM: A Survey on RL-Based Reasoning in Multimodal Large Language Models](https://mlsanity.com/paper/2504.21277/) (2025)
- [MMRC: A Large-Scale Benchmark for Understanding Multimodal Large Language Model in Real-World Conversation](https://mlsanity.com/paper/2502.11903/) (2025)
- [A Survey on Multimodal Benchmarks: In the Era of Large AI Models](https://mlsanity.com/paper/2409.18142/) (2024)
- [UniCorn: Towards Self-Improving Unified Multimodal Models through Self-Generated Supervision](https://mlsanity.com/paper/2601.03193/) (2026)
- [Benchmarking Multimodal Knowledge Conflict for Large Multimodal Models](https://mlsanity.com/paper/2505.19509/) (2025)
- [Can Visual Input Be Compressed? A Visual Token Compression Benchmark for Large Multimodal Models](https://mlsanity.com/paper/2511.02650/) (2025)
- [When Continue Learning Meets Multimodal Large Language Model: A Survey](https://mlsanity.com/paper/2503.01887/) (2025)
- [PairBench: Are Vision-Language Models Reliable at Comparing What They See?](https://mlsanity.com/paper/2502.15210/) (2025)
- [AVA-Bench: Atomic Visual Ability Benchmark for Vision Foundation Models](https://mlsanity.com/paper/2506.09082/) (2025)
- [Analyzing CLIP's Performance Limitations in Multi-Object Scenarios: A Controlled High-Resolution Study](https://mlsanity.com/paper/2502.19828/) (2025)
- [VOILA: Evaluation of MLLMs For Perceptual Understanding and Analogical Reasoning](https://mlsanity.com/paper/2503.00043/) (2025)
- [MV-Debate: Multi-view Agent Debate with Dynamic Reflection Gating for Multimodal Harmful Content Detection in Social Media](https://mlsanity.com/paper/2508.05557/) (2025)
- [PARC: A Quantitative Framework Uncovering the Symmetries within Vision Language Models](https://mlsanity.com/paper/2506.14808/) (2025)
- [MedBookVQA: A Systematic and Comprehensive Medical Benchmark Derived from Open-Access Book](https://mlsanity.com/paper/2506.00855/) (2025)
- [Breaking Bad Molecules: Are MLLMs Ready for Structure-Level Molecular Detoxification?](https://mlsanity.com/paper/2506.10912/) (2025)
- [MPCC: A Novel Benchmark for Multimodal Planning with Complex Constraints in Multimodal Large Language Models](https://mlsanity.com/paper/2507.23382/) (2025)
- [Multimodal LLMs See Sentiment](https://mlsanity.com/paper/2508.16873/) (2025)
- [Towards Scalable Web Accessibility Audit with MLLMs as Copilots](https://mlsanity.com/paper/2511.03471/) (2025)
- [Project MPG: towards a generalized performance benchmark for LLM capabilities](https://mlsanity.com/paper/2410.22368/) (2024)
- [A Lightweight Framework for Trigger-Guided LoRA-Based Self-Adaptation in LLMs](https://mlsanity.com/paper/2509.05385/) (2025)
- [MIDAS: Multi-Image Dispersion and Semantic Reconstruction for Jailbreaking MLLMs](https://mlsanity.com/paper/2603.00565/) (2026)
- [Granulon: Awakening Pixel-Level Visual Encoders with Adaptive Multi-Granularity Semantics for MLLM](https://mlsanity.com/paper/2603.08800/) (2026)
- [Role-RL: Online Long-Context Processing with Role Reinforcement Learning for Distinct LLMs in Their Optimal Roles](https://mlsanity.com/paper/2409.18014/) (2024)
- [Retrieval Enhanced Feedback via In-context Neural Error-book](https://mlsanity.com/paper/2508.16313/) (2025)
- [Evaluating Time Awareness and Cross-modal Active Perception of Large Models via 4D Escape Room Task](https://mlsanity.com/paper/2603.15467/) (2026)
- [AMIGO: Agentic Multi-Image Grounding Oracle Benchmark](https://mlsanity.com/paper/2603.28662/) (2026)
- [AgroTools: A Benchmark for Tool-Augmented Multimodal Agents in Agriculture](https://mlsanity.com/paper/2605.22366/) (2026)
- [Does VLA Even Know the Basics? Measuring Commonsense and World Knowledge Retention in Vision-Language-Action Models](https://mlsanity.com/paper/2606.19297/) (2026)
- [DiCoBench: Benchmarking Multi-Image Fine-Grained Perception via Differential and Commonality Visual Cues](https://mlsanity.com/paper/2606.26602/) (2026)

## Abstract (arXiv)

The ability to compare objects, scenes, or situations is crucial for effective decision-making and problem-solving in everyday life. For instance, comparing the freshness of apples enables better choices during grocery shopping while comparing sofa designs helps optimize the aesthetics of our living space. Despite its significance, the comparative capability is largely unexplored in artificial general intelligence (AGI). In this paper, we introduce MLLM-CompBench, a benchmark designed to evaluate the comparative reasoning capability of multimodal large language models (MLLMs). MLLM-CompBench mines and pairs images through visually oriented questions covering eight dimensions of relative comparison: visual attribute, existence, state, emotion, temporality, spatiality, quantity, and quality. We curate a collection of around 40K image pairs using metadata from diverse vision datasets and CLIP similarity scores. These image pairs span a broad array of visual domains, including animals, fashion, sports, and both outdoor and indoor scenes. The questions are carefully crafted to discern relative characteristics between two images and are labeled by human annotators for accuracy and relevance. We use MLLM-CompBench to evaluate recent MLLMs, including GPT-4V(ision), Gemini-Pro, and LLaVA-1.6. Our results reveal notable shortcomings in their comparative abilities. We believe MLLM-COMPBENCH not only sheds light on these limitations but also establishes a solid foundation for future enhancements in the comparative capability of MLLMs.