# What Users Leave Unsaid: Under-Specified Queries Limit Vision-Language Models

arXiv: https://arxiv.org/abs/2601.06165 · HTML: https://mlsanity.com/paper/2601.06165/

## Key facts

- Cited by 2 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 20 of its 30 references that are in the map.
- First posted on arXiv 7 Jan 2026.

## 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, 20 of 20)

- [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021)
- [HellaSwag: Can a Machine Really Finish Your Sentence?](https://mlsanity.com/paper/1905.07830/) (2019)
- [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)
- [InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency](https://mlsanity.com/paper/2508.18265/) (2025)
- [MM-Vet: Evaluating Large Multimodal Models for Integrated Capabilities](https://mlsanity.com/paper/2308.02490/) (2023)
- [MMMU-Pro: A More Robust Multi-discipline Multimodal Understanding Benchmark](https://mlsanity.com/paper/2409.02813/) (2024)
- [AmbigQA: Answering Ambiguous Open-domain Questions](https://mlsanity.com/paper/2004.10645/) (2020)
- [The BiGGen Bench: A Principled Benchmark for Fine-grained Evaluation of Language Models with Language Models](https://mlsanity.com/paper/2406.05761/) (2024)
- [KMMLU: Measuring Massive Multitask Language Understanding in Korean](https://mlsanity.com/paper/2402.11548/) (2024)
- [Building and Evaluating Open-Domain Dialogue Corpora with Clarifying Questions](https://mlsanity.com/paper/2109.05794/) (2021)
- [QuestBench: Can LLMs ask the right question to acquire information in reasoning tasks?](https://mlsanity.com/paper/2503.22674/) (2025)
- [ConvAI3: Generating Clarifying Questions for Open-Domain Dialogue Systems (ClariQ)](https://mlsanity.com/paper/2009.11352/) (2020)
- [HAE-RAE Bench: Evaluation of Korean Knowledge in Language Models](https://mlsanity.com/paper/2309.02706/) (2023)
- [A Survey on Asking Clarification Questions Datasets in Conversational Systems](https://mlsanity.com/paper/2305.15933/) (2023)
- [Evaluating Visual and Cultural Interpretation: The K-Viscuit Benchmark with Human-VLM Collaboration](https://mlsanity.com/paper/2406.16469/) (2024)
- [Teaching Vision-Language Models to Ask: Resolving Ambiguity in Visual Questions](https://mlsanity.com/paper/2507.13773/) (2025)
- [KRETA: A Benchmark for Korean Reading and Reasoning in Text-Rich VQA Attuned to Diverse Visual Contexts](https://mlsanity.com/paper/2508.19944/) (2025)
- [VARCO-VISION-2.0 Technical Report](https://mlsanity.com/paper/2509.10105/) (2025)
- [KOFFVQA: An Objectively Evaluated Free-form VQA Benchmark for Large Vision-Language Models in the Korean Language](https://mlsanity.com/paper/2503.23730/) (2025)

## Built on by (most cited first, 2 of 2)

- [KorMedMCQA-V: A Multimodal Benchmark for Evaluating Vision-Language Models on the Korean Medical Licensing Examination](https://mlsanity.com/paper/2602.13650/) (2026)
- [Text as Partial Constraint: Core-Residual Alignment for Robust Vision-Language Learning](https://mlsanity.com/paper/2607.03143/) (2026)

## Abstract (arXiv)

Current vision-language benchmarks predominantly feature well-structured questions with clear, explicit prompts. However, real user queries are often informal and underspecified. Users naturally leave much unsaid, relying on images to convey context. We introduce HAERAE-Vision, a benchmark of 653 real-world visual questions from Korean online communities (0.76% survival from 86K candidates), each paired with an explicit rewrite, yielding 1,306 query variants in total. Evaluating 39 VLMs, we find that even state-of-the-art models (GPT-5, Gemini 2.5 Pro) achieve under 50% on the original queries. Crucially, query explicitation alone yields 8 to 22 point improvements, with smaller models benefiting most. We further show that even with web search, under-specified queries underperform explicit queries without search, revealing that current retrieval cannot compensate for what users leave unsaid. Our findings demonstrate that a substantial portion of VLM difficulty stem from natural query under-specification instead of model capability, highlighting a critical gap between benchmark evaluation and real-world deployment.