# Benchmarking and Mitigating MCQA Selection Bias of Large Vision-Language Models

arXiv: https://arxiv.org/abs/2509.16805 · HTML: https://mlsanity.com/paper/2509.16805/

## Key facts

- Cited by 5 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 18 of its 42 references that are in the map.
- First posted on arXiv 20 Sep 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, 18 of 18)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Improved Baselines with Visual Instruction Tuning](https://mlsanity.com/paper/2310.03744/) (2023)
- [InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning](https://mlsanity.com/paper/2305.06500/) (2023)
- [Fine-Grained Visual Classification of Aircraft](https://mlsanity.com/paper/1306.5151/) (2013)
- [Large Language Models are not Fair Evaluators](https://mlsanity.com/paper/2305.17926/) (2023)
- [Janus-Pro: Unified Multimodal Understanding and Generation with Data and Model Scaling](https://mlsanity.com/paper/2501.17811/) (2025)
- [Large Language Models Are Not Robust Multiple Choice Selectors](https://mlsanity.com/paper/2309.03882/) (2023)
- [Leveraging Large Language Models for Multiple Choice Question Answering](https://mlsanity.com/paper/2210.12353/) (2022)
- [Are Gender-Neutral Queries Really Gender-Neutral? Mitigating Gender Bias in Image Search](https://mlsanity.com/paper/2109.05433/) (2021)
- [Can multiple-choice questions really be useful in detecting the abilities of LLMs?](https://mlsanity.com/paper/2403.17752/) (2024)
- [MUTANT: A Training Paradigm for Out-of-Distribution Generalization in Visual Question Answering](https://mlsanity.com/paper/2009.08566/) (2020)
- [Which of These Best Describes Multiple Choice Evaluation with LLMs? A) Forced B) Flawed C) Fixable D) All of the Above](https://mlsanity.com/paper/2502.14127/) (2025)
- [LLMs May Perform MCQA by Selecting the Least Incorrect Option](https://mlsanity.com/paper/2402.01349/) (2024)
- [UniBias: Unveiling and Mitigating LLM Bias through Internal Attention and FFN Manipulation](https://mlsanity.com/paper/2405.20612/) (2024)
- [Mitigating Biases for Instruction-following Language Models via Bias Neurons Elimination](https://mlsanity.com/paper/2311.09627/) (2023)
- [Strengthened Symbol Binding Makes Large Language Models Reliable Multiple-Choice Selectors](https://mlsanity.com/paper/2406.01026/) (2024)
- [Discovering Bias in Latent Space: An Unsupervised Debiasing Approach](https://mlsanity.com/paper/2406.03631/) (2024)

## Built on by (most cited first, 5 of 5)

- [PEARL: Personalized Streaming Video Understanding Model](https://mlsanity.com/paper/2603.20422/) (2026)
- [BUZZY: Contrastive Scoring to Mitigate Text-Induced Bias in Multimodal Multiple-Choice QA](https://mlsanity.com/paper/2603.28026/) (2026)
- [CounterCount: A Diagnostic Framework for Counting Bias in Vision Language Models](https://mlsanity.com/paper/2605.17826/) (2026)
- [Candidate-Expanding Routing with Permutation-Stabilized Experts for Mixed-Format Medical VQA](https://mlsanity.com/paper/2609.00959/) (2026)
- [What Looks Like a Capability Limit in Vision-Language Models Is a Readout Limit](https://mlsanity.com/paper/2609.27408/) (2026)

## Abstract (arXiv)

Large Vision-Language Models (LVLMs) have achieved strong performance on vision-language tasks, particularly Visual Question Answering (VQA). While prior work has explored unimodal biases in VQA, the problem of selection bias in Multiple-Choice Question Answering (MCQA), where models may favor specific option tokens (e.g., "A") or positions, remains underexplored. In this paper, we investigate both the presence and nature of selection bias in LVLMs through fine-grained MCQA benchmarks spanning easy, medium, and hard difficulty levels, defined by the semantic similarity of the options. We further propose an inference-time logit-level debiasing method that estimates an ensemble bias vector from general and contextual prompts and applies confidence-adaptive corrections to the model's output. Our method mitigates bias without retraining and is compatible with frozen LVLMs. Extensive experiments across several state-of-the-art models reveal consistent selection biases that intensify with task difficulty, and show that our mitigation approach significantly reduces bias while improving accuracy in challenging settings. This work offers new insights into the limitations of LVLMs in MCQA and presents a practical approach to improve their robustness in fine-grained visual reasoning. Datasets and code are available at: https://github.com/Atabuzzaman/Selection-Bias-of-LVLMs