# XYBench: Can LLMs Respond Pragmatically to Queries with Misconceptions?

arXiv: https://arxiv.org/abs/2609.06842 · HTML: https://mlsanity.com/paper/2609.06842/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 46 references that are in the map.
- First posted on arXiv 6 Sep 2026.

## Builds on (most cited first, 19 of 19)

- [Know What You Don't Know: Unanswerable Questions for SQuAD](https://mlsanity.com/paper/1806.03822/) (2018)
- [Length-Controlled AlpacaEval: A Simple Way to Debias Automatic Evaluators](https://mlsanity.com/paper/2404.04475/) (2024)
- [LLMs Get Lost In Multi-Turn Conversation](https://mlsanity.com/paper/2505.06120/) (2025)
- [AmbigQA: Answering Ambiguous Open-domain Questions](https://mlsanity.com/paper/2004.10645/) (2020)
- [QuAC : Question Answering in Context](https://mlsanity.com/paper/1808.07036/) (2018)
- [Neural Theory-of-Mind? On the Limits of Social Intelligence in Large LMs](https://mlsanity.com/paper/2210.13312/) (2022)
- [WikiHow: A Large Scale Text Summarization Dataset](https://mlsanity.com/paper/1810.09305/) (2018)
- [CREPE: Open-Domain Question Answering with False Presuppositions](https://mlsanity.com/paper/2211.17257/) (2022)
- [Challenges in Information-Seeking QA: Unanswerable Questions and Paragraph Retrieval](https://mlsanity.com/paper/2010.11915/) (2020)
- [Which Linguist Invented the Lightbulb? Presupposition Verification for Question-Answering](https://mlsanity.com/paper/2101.00391/) (2021)
- [Ambig-SWE: Interactive Agents to Overcome Underspecificity in Software Engineering](https://mlsanity.com/paper/2502.13069/) (2025)
- [Human-Centered Design Recommendations for LLM-as-a-Judge](https://mlsanity.com/paper/2407.03479/) (2024)
- [Be Friendly, Not Friends: How LLM Sycophancy Shapes User Trust](https://mlsanity.com/paper/2502.10844/) (2025)
- [Learning to Make MISTAKEs: Modeling Incorrect Student Thinking And Key Errors](https://mlsanity.com/paper/2510.11502/) (2025)
- [Evaluating Large Language Models for Health-related Queries with Presuppositions](https://mlsanity.com/paper/2312.08800/) (2023)
- [Can LLMs Ground when they (Don't) Know: A Study on Direct and Loaded Political Questions](https://mlsanity.com/paper/2506.08952/) (2025)
- [A Taxonomy of Ambiguity Types for NLP](https://mlsanity.com/paper/2403.14072/) (2024)
- [Superlatives in Context: Modeling the Implicit Semantics of Superlatives](https://mlsanity.com/paper/2405.20967/) (2024)
- [Taking Action Towards Graceful Interaction: The Effects of Performing Actions on Modelling Policies for Instruction Clarification Requests](https://mlsanity.com/paper/2401.17039/) (2024)

## Built on by (most cited first, 1 of 1)

- [XYEval: Agents say yes to bad advice](https://mlsanity.com/paper/2609.23939/) (2026)

## Abstract (arXiv)

When non-expert users ask LLMs for assistance, their queries can often have misconceptions (e.g., "How do I parse XML with regex?"). In such cases, often referred to as the XY-problem, LLMs must identify the misconception ("regex are fragile") and meaningfully direct the user toward a pragmatic solution that will address the root problem implicit in the request ("use an XML parser"). We introduce XYBench, a benchmark of 8,115 such queries, drawn from technical (StackOverflow/StackExchange) and everyday (WikiHow and a manually-curated subset) domains. We design an evaluation paradigm that assesses model responses along three criteria grounded in cooperative response theory: (a) presence and (b) emphasis on pragmatic solutions, and (c) identification of misconceptions. Our experiments show that even the strongest LLMs predominantly answer the literal request (0.75--0.92) and far less often the intended one (0.33--0.71), while substantially lagging behind humans at identifying misconceptions (at most 63% vs. 79--90%). Further, models overwhelmingly prefer pragmatic responses in a multiple choice setting yet consistently fail to generate them. Oracle ablation experiments show that providing explicit user intent at generation time helps; however a large gap remains, suggesting pragmatic redirection is a fundamentally underdeveloped capability in current LLMs.