测试大模型能否识别并纠正用户提问中的错误假设。
Evaluating Reasoning Models for Queries with Presuppositions
- 构建含不同假设强度的健康、科学类问题数据集
- 推理模型准确率仅提升2-11%，仍有26-42%错误假设未被挑战
- 模型表现受假设表达强度影响，越强烈越难纠正
数百万用户依赖AI模型获取信息，但许多查询隐含事实错误的前提。以往研究发现大型语言模型（LLMs）常无法质疑这些错误假设，甚至强化用户偏见。随着模型推理能力提升，我们重新评估大型推理模型（LRMs）是否能正确处理此类问题。构建涵盖健康、科学及通用知识领域、具有不同程度预设的查询数据集，评估多个广泛部署的模型。结果显示，相比非推理模型，推理模型准确率提升2%-11%，但仍无法挑战26%-42%的错误前提。此外，模型对假设表达强度敏感，越强烈的预设越难被纠正。
原文摘要 · Abstract (English)
Millions of users turn to AI models for their information needs. It is conceivable that a large number of user queries contain assumptions that may be factually inaccurate. Prior work notes that large language models (LLMs) often fail to challenge such erroneous assumptions, and can reinforce users' misinformed opinions. However, given the recent advances, especially in model's reasoning capabilities, we revisit whether large reasoning models (LRMs) can reason about the underlying assumptions and respond to user queries appropriately. We construct queries with varying degrees of presuppositions spanning health, science, and general knowledge, and use it to evaluate several widely-deployed models When compared to non-reasoning models, we find that reasoning models achieve a slightly higher accuracy (2-11%), but they still fail to challenge a large fraction (26-42%) of false presuppositions. Further, reasoning models remain susceptible to how strongly the presupposition is expressed.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。