# Evaluating Cross-Modal Reasoning Ability and Problem Characteristics with Multimodal Item Response Theory

arXiv: https://arxiv.org/abs/2603.02663 · HTML: https://mlsanity.com/paper/2603.02663/

## 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 26 of its 39 references that are in the map.
- First posted on arXiv 3 Mar 2026.

## Placed in

- LLM reasoning / [Multimodal reasoning](https://mlsanity.com/lr/multimodal-reasoning/)
- Multimodal LLMs / vision-language models / [MLLM training and visual instruction tuning](https://mlsanity.com/mllm/mllm-training-and-visual-instruction-tuning/)

## Builds on (most cited first, 26 of 26)

- [Making the V in VQA Matter: Elevating the Role of Image Understanding in Visual Question Answering](https://mlsanity.com/paper/1612.00837/) (2016)
- [MMBench: Is Your Multi-modal Model an All-around Player?](https://mlsanity.com/paper/2307.06281/) (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)
- [Are We on the Right Way for Evaluating Large Vision-Language Models?](https://mlsanity.com/paper/2403.20330/) (2024)
- [A Survey on Multimodal Large Language Models](https://mlsanity.com/paper/2306.13549/) (2023)
- [VizWiz Grand Challenge: Answering Visual Questions from Blind People](https://mlsanity.com/paper/1802.08218/) (2018)
- [Vision-Language Models for Vision Tasks: A Survey](https://mlsanity.com/paper/2304.00685/) (2023)
- [MMMU-Pro: A More Robust Multi-discipline Multimodal Understanding Benchmark](https://mlsanity.com/paper/2409.02813/) (2024)
- [MiniMax-01: Scaling Foundation Models with Lightning Attention](https://mlsanity.com/paper/2501.08313/) (2025)
- [Can MLLMs Reason in Multimodality? EMMA: An Enhanced MultiModal ReAsoning Benchmark](https://mlsanity.com/paper/2501.05444/) (2025)
- [Cross-Modal Implicit Relation Reasoning and Aligning for Text-to-Image Person Retrieval](https://mlsanity.com/paper/2303.12501/) (2023)
- [Anchor Points: Benchmarking Models with Much Fewer Examples](https://mlsanity.com/paper/2309.08638/) (2023)
- [metabench -- A Sparse Benchmark of Reasoning and Knowledge in Large Language Models](https://mlsanity.com/paper/2407.12844/) (2024)
- [Active Evaluation Acquisition for Efficient LLM Benchmarking](https://mlsanity.com/paper/2410.05952/) (2024)
- [How Predictable Are Large Language Model Capabilities? A Case Study on BIG-bench](https://mlsanity.com/paper/2305.14947/) (2023)
- [Automated Evaluation of Retrieval-Augmented Language Models with Task-Specific Exam Generation](https://mlsanity.com/paper/2405.13622/) (2024)
- [Dynamic Multimodal Evaluation with Flexible Complexity by Vision-Language Bootstrapping](https://mlsanity.com/paper/2410.08695/) (2024)
- [Raising the Bar: Investigating the Values of Large Language Models via Generative Evolving Testing](https://mlsanity.com/paper/2406.14230/) (2024)
- [AcTracer: Active Testing of Large Language Model via Multi-Stage Sampling](https://mlsanity.com/paper/2408.03573/) (2024)
- [LiveXiv -- A Multi-Modal Live Benchmark Based on Arxiv Papers Content](https://mlsanity.com/paper/2410.10783/) (2024)
- [VQA Therapy: Exploring Answer Differences by Visually Grounding Answers](https://mlsanity.com/paper/2308.11662/) (2023)
- [Data Efficient Evaluation of Large Language Models and Text-to-Image Models via Adaptive Sampling](https://mlsanity.com/paper/2406.15527/) (2024)
- [MAC: A Live Benchmark for Multimodal Large Language Models in Scientific Understanding](https://mlsanity.com/paper/2508.15802/) (2025)
- [FlashEval: Towards Fast and Accurate Evaluation of Text-to-image Diffusion Generative Models](https://mlsanity.com/paper/2403.16379/) (2024)
- [CCHall: A Novel Benchmark for Joint Cross-Lingual and Cross-Modal Hallucinations Detection in Large Language Models](https://mlsanity.com/paper/2505.19108/) (2025)

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

- [Same Evidence, Different Answer: Auditing Order Sensitivity in Multimodal Large Language Models](https://mlsanity.com/paper/2606.26079/) (2026)
- [Hierarchical Compression of Vision-Language Model Benchmarks](https://mlsanity.com/paper/2609.37515/) (2026)
- [Are Multimodal Benchmarks Really Useful? Item-Level Multimodal Benchmark Diagnosis via Structure-Response Co-Calibration](https://mlsanity.com/paper/2610.14542/) (2026)

## Abstract (arXiv)

Multimodal Large Language Models (MLLMs) have recently emerged as general architectures capable of reasoning over diverse modalities. Benchmarks for MLLMs should measure their ability for cross-modal integration. However, current benchmarks are filled with shortcut questions, which can be solved using only a single modality, thereby yielding unreliable rankings. For example, in vision-language cases, we can find the correct answer without either the image or the text. These low-quality questions unnecessarily increase the size and computational requirements of benchmarks. We introduce a multi-modal and multidimensional item response theory framework (M3IRT) that extends classical IRT by decomposing both model ability and item difficulty into image-only, text-only, and cross-modal components. M3IRT estimates cross-modal ability of MLLMs and each question's cross-modal difficulty, enabling compact, high-quality subsets that better reflect multimodal reasoning. Across 24 VLMs on three benchmarks, M3IRT prioritizes genuinely cross-modal questions over shortcuts and preserves ranking fidelity even when 50% of items are artificially generated low-quality questions, thereby reducing evaluation cost while improving reliability. M3IRT thus offers a practical tool for assessing cross-modal reasoning and refining multimodal benchmarks.