# CompareBench: A Benchmark for Visual Comparison Reasoning in Vision-Language Models

arXiv: https://arxiv.org/abs/2509.22737 · HTML: https://mlsanity.com/paper/2509.22737/

## Key facts

- Cited by 3 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 16 references that are in the map.
- First posted on arXiv 25 Sep 2025.

## 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)

- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [MMBench: Is Your Multi-modal Model an All-around Player?](https://mlsanity.com/paper/2307.06281/) (2023)
- [Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering](https://mlsanity.com/paper/2209.09513/) (2022)
- [CLEVR: A Diagnostic Dataset for Compositional Language and Elementary Visual Reasoning](https://mlsanity.com/paper/1612.06890/) (2016)
- [MM-Vet: Evaluating Large Multimodal Models for Integrated Capabilities](https://mlsanity.com/paper/2308.02490/) (2023)
- [BLINK: Multimodal Large Language Models Can See but Not Perceive](https://mlsanity.com/paper/2404.12390/) (2024)
- [Teaching CLIP to Count to Ten](https://mlsanity.com/paper/2302.12066/) (2023)
- [Drone-based Object Counting by Spatially Regularized Regional Proposal Network](https://mlsanity.com/paper/1707.05972/) (2017)
- [TallyQA: Answering Complex Counting Questions](https://mlsanity.com/paper/1810.12440/) (2018)
- [Learning To Count Everything](https://mlsanity.com/paper/2104.08391/) (2021)
- [SimpleVQA: Multimodal Factuality Evaluation for Multimodal Large Language Models](https://mlsanity.com/paper/2502.13059/) (2025)
- [Open-world Text-specified Object Counting](https://mlsanity.com/paper/2306.01851/) (2023)

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

- [Thinking with Visual Grounding](https://mlsanity.com/paper/2606.16122/) (2026)
- [V-REX: Benchmarking Exploratory Visual Reasoning via Chain-of-Questions](https://mlsanity.com/paper/2512.11995/) (2025)
- [Seeing Time: Benchmarking Chronological Reasoning and Shortcut Biases in Vision-Language Models](https://mlsanity.com/paper/2606.05702/) (2026)

## Abstract (arXiv)

Visual comparison reasoning is a fundamental capability of vision-language models (VLMs), covering judgments of object quantity, geometric dimensions, spatial relations, and temporal order. Yet existing benchmarks rarely isolate comparison as a reasoning axis, leaving it unclear whether models can reliably perform comparative visual judgments. We introduce a benchmark suite organized around three top-level resources: TallyBench, a 2,000-image object counting benchmark; OmniCaps, a 716-image caption and tag resource; and CompareBench, a 1,200-QA visual comparison benchmark. CompareBench contains four sub-benchmarks spanning quantity, geometric, spatial, and temporal comparison, with the temporal component unifying historical scenes, landmarks, and public figures. Evaluating nine closed-source model routes from Anthropic, Google, and OpenAI on TallyBench and CompareBench reveals strong overall performance but persistent failures in counting, spatial reasoning, geometric comparison, and temporal ordering. These results show that visual comparison remains a systematic weakness of current VLMs and establish CompareBench as a focused benchmark for multimodal reasoning evaluation. All data, code, and prompts will be released at https://github.com/caijie0620/CompareBench.