# MoHoBench: Assessing Honesty of Multimodal Large Language Models via Unanswerable Visual Questions

arXiv: https://arxiv.org/abs/2507.21503 · HTML: https://mlsanity.com/paper/2507.21503/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Placed in MLLM training and visual instruction tuning, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 33 of its 69 references that are in the map.
- First posted on arXiv 29 Jul 2025.

## Placed in

- 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, 30 of 33)

- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [SQuAD: 100,000+ Questions for Machine Comprehension of Text](https://mlsanity.com/paper/1606.05250/) (2016)
- [SimCSE: Simple Contrastive Learning of Sentence Embeddings](https://mlsanity.com/paper/2104.08821/) (2021)
- [Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone](https://mlsanity.com/paper/2404.14219/) (2024)
- [LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models](https://mlsanity.com/paper/2403.13372/) (2024)
- [A Survey on LLM-as-a-Judge](https://mlsanity.com/paper/2411.15594/) (2024)
- [A General Language Assistant as a Laboratory for Alignment](https://mlsanity.com/paper/2112.00861/) (2021)
- [SimPO: Simple Preference Optimization with a Reference-Free Reward](https://mlsanity.com/paper/2405.14734/) (2024)
- [Teaching Models to Express Their Uncertainty in Words](https://mlsanity.com/paper/2205.14334/) (2022)
- [A Survey on In-context Learning](https://mlsanity.com/paper/2301.00234/) (2022)
- [Aligning Large Multimodal Models with Factually Augmented RLHF](https://mlsanity.com/paper/2309.14525/) (2023)
- [Object Hallucination in Image Captioning](https://mlsanity.com/paper/1809.02156/) (2018)
- [DeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understanding](https://mlsanity.com/paper/2412.10302/) (2024)
- [ORPO: Monolithic Preference Optimization without Reference Model](https://mlsanity.com/paper/2403.07691/) (2024)
- [Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAs](https://mlsanity.com/paper/2503.01743/) (2025)
- [LiveBench: A Challenging, Contamination-Limited LLM Benchmark](https://mlsanity.com/paper/2406.19314/) (2024)
- [Enhancing the Reasoning Ability of Multimodal Large Language Models via Mixed Preference Optimization](https://mlsanity.com/paper/2411.10442/) (2024)
- [LLaVA-Critic: Learning to Evaluate Multimodal Models](https://mlsanity.com/paper/2410.02712/) (2024)
- [Silkie: Preference Distillation for Large Visual Language Models](https://mlsanity.com/paper/2312.10665/) (2023)
- [Truthful AI: Developing and governing AI that does not lie](https://mlsanity.com/paper/2110.06674/) (2021)
- [How Many Unicorns Are in This Image? A Safety Evaluation Benchmark for Vision LLMs](https://mlsanity.com/paper/2311.16101/) (2023)
- [Reliable Visual Question Answering: Abstain Rather Than Answer Incorrectly](https://mlsanity.com/paper/2204.13631/) (2022)
- [MM-RLHF: The Next Step Forward in Multimodal LLM Alignment](https://mlsanity.com/paper/2502.10391/) (2025)
- [CIEM: Contrastive Instruction Evaluation Method for Better Instruction Tuning](https://mlsanity.com/paper/2309.02301/) (2023)
- [VLFeedback: A Large-Scale AI Feedback Dataset for Large Vision-Language Models Alignment](https://mlsanity.com/paper/2410.09421/) (2024)
- [Examining LLMs' Uncertainty Expression Towards Questions Outside Parametric Knowledge](https://mlsanity.com/paper/2311.09731/) (2023)
- [ToViLaG: Your Visual-Language Generative Model is Also An Evildoer](https://mlsanity.com/paper/2312.11523/) (2023)

## Built on by (most cited first, 11 of 11)

- [Inflated Excellence or True Performance? Rethinking Medical Diagnostic Benchmarks with Dynamic Evaluation](https://mlsanity.com/paper/2510.09275/) (2025)
- [Counterfactual Reasoning for Steerable Pluralistic Value Alignment of Large Language Models](https://mlsanity.com/paper/2510.18526/) (2025)
- [HypeMed: Enhancing Medication Recommendations with Hypergraph-Based Patient Relationships](https://mlsanity.com/paper/2603.18459/) (2026)
- [Senses Wide Shut: A Representation-Action Gap in Omnimodal LLMs](https://mlsanity.com/paper/2605.13737/) (2026)
- [MMOOC: A Comprehensive Benchmark for Out-of-Context Evaluation in Multimodal Large Language Models](https://mlsanity.com/paper/2607.27637/) (2026)
- [Controllable Affective Generation via Latent Vector Steering](https://mlsanity.com/paper/2608.25569/) (2026)
- [MMPCBench: Benchmarking Multimodal Large Language Models on Proactive Critique of Flawed Inputs](https://mlsanity.com/paper/2608.29286/) (2026)
- [Knowing What Not to Answer: Selective Non-Compliance in Vision-Language Models](https://mlsanity.com/paper/2609.04720/) (2026)
- [From Static Personal Values to Contextualized Personalization: Bayesian Personalized Value Alignment for LLMs](https://mlsanity.com/paper/2609.28942/) (2026)
- [When Does an Image Determine the Answer? Benchmarking Visual Answerability across Charts and Scenes](https://mlsanity.com/paper/2609.34480/) (2026)
- [Visual Dependence Does Not Imply Answerability: A Systematic Analysis of Answerability Signals in Frozen InstructBLIP](https://mlsanity.com/paper/2610.17902/) (2026)

## Abstract (arXiv)

Recently Multimodal Large Language Models (MLLMs) have achieved considerable advancements in vision-language tasks, yet produce potentially harmful or untrustworthy content. Despite substantial work investigating the trustworthiness of language models, MMLMs' capability to act honestly, especially when faced with visually unanswerable questions, remains largely underexplored. This work presents the first systematic assessment of honesty behaviors across various MLLMs. We ground honesty in models' response behaviors to unanswerable visual questions, define four representative types of such questions, and construct MoHoBench, a large-scale MMLM honest benchmark, consisting of 12k+ visual question samples, whose quality is guaranteed by multi-stage filtering and human verification. Using MoHoBench, we benchmarked the honesty of 28 popular MMLMs and conducted a comprehensive analysis. Our findings show that: (1) most models fail to appropriately refuse to answer when necessary, and (2) MMLMs' honesty is not solely a language modeling issue, but is deeply influenced by visual information, necessitating the development of dedicated methods for multimodal honesty alignment. Therefore, we implemented initial alignment methods using supervised and preference learning to improve honesty behavior, providing a foundation for future work on trustworthy MLLMs. Our data and code can be found at https://github.com/yanxuzhu/MoHoBench.