AbstentionBench: Reasoning LLMs Fail on Unanswerable Questions

• URL: http://arxiv.org/abs/2506.09038v1
• Date: Tue, 10 Jun 2025 17:57:30 GMT
• Title: AbstentionBench: Reasoning LLMs Fail on Unanswerable Questions
• Authors: Polina Kirichenko, Mark Ibrahim, Kamalika Chaudhuri, Samuel J. Bell,
• Abstract summary: AbstentionBench is a benchmark for holistically evaluating abstention across 20 diverse datasets.<n>We find that reasoning fine-tuning degrades abstention even for math and science domains.
• Score: 32.871820908561936
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: For Large Language Models (LLMs) to be reliably deployed in both everyday and high-stakes domains, knowing when not to answer is equally critical as answering correctly. Real-world user queries, which can be underspecified, ill-posed, or fundamentally unanswerable, require LLMs to reason about uncertainty and selectively abstain -- i.e., refuse to answer definitively. However, abstention remains understudied, without a systematic evaluation framework for modern LLMs. In this work, we introduce AbstentionBench, a large-scale benchmark for holistically evaluating abstention across 20 diverse datasets, including questions with unknown answers, underspecification, false premises, subjective interpretations, and outdated information. Evaluating 20 frontier LLMs reveals abstention is an unsolved problem, and one where scaling models is of little use. While recent reasoning LLMs have shown impressive results in complex problem solving, surprisingly, we find that reasoning fine-tuning degrades abstention (by $24\%$ on average), even for math and science domains on which reasoning models are explicitly trained. We find that while a carefully crafted system prompt can boost abstention in practice, it does not resolve models' fundamental inability to reason about uncertainty. We release AbstentionBench to foster research into advancing LLM reliability.

Related papers

• WakenLLM: Evaluating Reasoning Potential and Stability in LLMs via Fine-Grained Benchmarking [14.76224690767612]
  Large Language Models (LLMs) frequently output the label Unknown in reasoning tasks.<n>We introduce WakenLLM, a framework that quantifies the portion of Unknown output attributable to model incapacity.
  arXiv  Detail & Related papers  (2025-07-22T03:21:48Z)
• ReliableMath: Benchmark of Reliable Mathematical Reasoning on Large Language Models [70.33764118171463]
  Large Language Models (LLMs) tend to fabricate unreliable responses when confronted with problems that are unsolvable or beyond their capability.<n>We develop a ReliableMath dataset which incorporates open-source solvable problems and high-quality unsolvable problems.<n>LLMs fail to directly identify unsolvable problems and always generate fabricated responses.
  arXiv  Detail & Related papers  (2025-07-03T19:19:44Z)
• Reasoning Capabilities and Invariability of Large Language Models [49.23570751696334]
  We aim to provide a comprehensive analysis of Large Language Models' reasoning competence.<n>We introduce a new benchmark dataset with a series of simple reasoning questions demanding shallow logical reasoning.<n>An empirical analysis involving zero-shot and few-shot prompting across 24 LLMs of different sizes reveals that, while LLMs with over 70 billion parameters perform better in the zero-shot setting, there is still a large room for improvement.
  arXiv  Detail & Related papers  (2025-05-01T18:12:30Z)
• None of the Above, Less of the Right: Parallel Patterns between Humans and LLMs on Multi-Choice Questions Answering [23.176280217718002]
  Multiple-choice exam questions with "None of the above" (NA) options have been extensively studied in educational testing.<n>We examine how NA options affect model performance and confidence calibration.<n>Our results highlight important implications for benchmark design and raise questions about LLMs' ability to handle uncertainty in real-world applications.
  arXiv  Detail & Related papers  (2025-03-03T13:55:29Z)
• Exploring Language Model Generalization in Low-Resource Extractive QA [57.14068405860034]
  We investigate Extractive Question Answering (EQA) with Large Language Models (LLMs) under domain drift.<n>We devise a series of experiments to explain the performance gap empirically.
  arXiv  Detail & Related papers  (2024-09-27T05:06:43Z)
• Understanding the Relationship between Prompts and Response Uncertainty in Large Language Models [55.332004960574004]
  Large language models (LLMs) are widely used in decision-making, but their reliability, especially in critical tasks like healthcare, is not well-established.<n>This paper investigates how the uncertainty of responses generated by LLMs relates to the information provided in the input prompt.<n>We propose a prompt-response concept model that explains how LLMs generate responses and helps understand the relationship between prompts and response uncertainty.
  arXiv  Detail & Related papers  (2024-07-20T11:19:58Z)
• Semantic Density: Uncertainty Quantification for Large Language Models through Confidence Measurement in Semantic Space [14.715989394285238]
  Existing Large Language Models (LLMs) do not have an inherent functionality to provide the users with an uncertainty/confidence metric for each response it generates. A new framework is proposed in this paper to address these issues. Semantic density extracts uncertainty/confidence information for each response from a probability distribution perspective in semantic space.
  arXiv  Detail & Related papers  (2024-05-22T17:13:49Z)
• Think Twice Before Trusting: Self-Detection for Large Language Models through Comprehensive Answer Reflection [90.71323430635593]
  We propose a novel self-detection paradigm that considers the comprehensive answer space beyond LLM-generated answers. Building upon this paradigm, we introduce a two-step framework, which firstly instructs LLM to reflect and provide justifications for each candidate answer. This framework can be seamlessly integrated with existing approaches for superior self-detection.
  arXiv  Detail & Related papers  (2024-03-15T02:38:26Z)
• Examining LLMs' Uncertainty Expression Towards Questions Outside Parametric Knowledge [35.067234242461545]
  Large language models (LLMs) express uncertainty in situations where they lack sufficient parametric knowledge to generate reasonable responses. This work aims to systematically investigate LLMs' behaviors in such situations, emphasizing the trade-off between honesty and helpfulness.
  arXiv  Detail & Related papers  (2023-11-16T10:02:40Z)
• Are Large Language Models Really Robust to Word-Level Perturbations? [68.60618778027694]
  We propose a novel rational evaluation approach that leverages pre-trained reward models as diagnostic tools. Longer conversations manifest the comprehensive grasp of language models in terms of their proficiency in understanding questions. Our results demonstrate that LLMs frequently exhibit vulnerability to word-level perturbations that are commonplace in daily language usage.
  arXiv  Detail & Related papers  (2023-09-20T09:23:46Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.