# On the robustness of multimodal language model towards distractions

arXiv: https://arxiv.org/abs/2502.09818 · HTML: https://mlsanity.com/paper/2502.09818/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Placed in VLM benchmarks and visual reasoning, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 9 of its 36 references that are in the map.
- First posted on arXiv 13 Feb 2025.

## Placed in

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

## Builds on (most cited first, 9 of 9)

- [MMBench: Is Your Multi-modal Model an All-around Player?](https://mlsanity.com/paper/2307.06281/) (2023)
- [MME: A Comprehensive Evaluation Benchmark for Multimodal Large Language Models](https://mlsanity.com/paper/2306.13394/) (2023)
- [MM-Vet: Evaluating Large Multimodal Models for Integrated Capabilities](https://mlsanity.com/paper/2308.02490/) (2023)
- [How Far Are We to GPT-4V? Closing the Gap to Commercial Multimodal Models with Open-Source Suites](https://mlsanity.com/paper/2404.16821/) (2024)
- [LISA: Reasoning Segmentation via Large Language Model](https://mlsanity.com/paper/2308.00692/) (2023)
- [HallusionBench: An Advanced Diagnostic Suite for Entangled Language Hallucination and Visual Illusion in Large Vision-Language Models](https://mlsanity.com/paper/2310.14566/) (2023)
- [How Robust is Google's Bard to Adversarial Image Attacks?](https://mlsanity.com/paper/2309.11751/) (2023)
- [HALC: Object Hallucination Reduction via Adaptive Focal-Contrast Decoding](https://mlsanity.com/paper/2403.00425/) (2024)
- [Large Multilingual Models Pivot Zero-Shot Multimodal Learning across Languages](https://mlsanity.com/paper/2308.12038/) (2023)

## Built on by (most cited first, 9 of 9)

- [When Audio and Text Disagree: Revealing Text Bias in Large Audio-Language Models](https://mlsanity.com/paper/2508.15407/) (2025)
- [Knowing the Answer Isn't Enough: Fixing Reasoning Path Failures in LVLMs](https://mlsanity.com/paper/2512.06258/) (2025)
- [When Silence Matters: The Impact of Irrelevant Audio on Text Reasoning in Large Audio-Language Models](https://mlsanity.com/paper/2510.00626/) (2025)
- [Can Large Multimodal Models Actively Recognize Faulty Inputs? A Systematic Evaluation Framework of Their Input Scrutiny Ability](https://mlsanity.com/paper/2508.04017/) (2025)
- [Red-teaming the Multimodal Reasoning: Jailbreaking Vision-Language Models via Cross-modal Entanglement Attacks](https://mlsanity.com/paper/2602.10148/) (2026)
- [NavTrust: Benchmarking Trustworthiness for Embodied Navigation](https://mlsanity.com/paper/2603.19229/) (2026)
- [Dynamic Adversarial Reinforcement Learning for Robust Multimodal Large Language Models](https://mlsanity.com/paper/2602.22227/) (2026)
- [The Effects of Visual Priming on Cooperative Behavior in Vision-Language Models](https://mlsanity.com/paper/2604.27953/) (2026)
- [Belief Without Behavior: Measuring the Translation of Theory of Mind into Coordinated Social Action in Vision-Language Models](https://mlsanity.com/paper/2608.20975/) (2026)

## Abstract (arXiv)

Although vision-language models (VLMs) have achieved significant success in various applications such as visual question answering, their resilience to prompt variations remains an under-explored area. Understanding how distractions affect VLMs is crucial for improving their real-world applicability, as inputs could have noisy and irrelevant information in many practical scenarios. This paper aims to assess the robustness of VLMs against both visual and textual distractions in the context of science question answering. Built on the ScienceQA dataset, we developed a new benchmark that introduces distractions in both the visual and textual contexts to evaluate the reasoning capacity of VLMs amid these distractions. Our findings reveal that most-of-the-art VLMs, including GPT-4, are vulnerable to various types of distractions, experiencing noticeable degradation in reasoning capabilities when confronted with distractions. Notably, models such as InternVL2 demonstrate a higher degree of robustness to these distractions. We also found that models exhibit greater sensitivity to textual distractions than visual ones. Additionally, we explored various mitigation strategies, such as prompt engineering, to counteract the impact of distractions. While these strategies improved solution accuracy, our analysis shows that there remain significant opportunities for improvement.