测试大模型是否真会逻辑推理，发现它们更依赖提示线索而非真正思考。
Reasoning or a Semblance of it? A Diagnostic Study of Transitive Reasoning in LLMs
- 通过控制关键词重叠、预训练知识和命名实体，隔离推理机制
- 两个模型都依赖提示线索，但Flan-T5对干扰更不敏感
- 结果暗示微调数据可能让模型习得真正的传递性推理能力
在推理基准上评估大语言模型（LLMs）的表现，显示其能解决复合型问题。然而，尚不清楚这些模型是进行真实逻辑推理，还是仅依赖隐含线索生成答案。本文通过操控两个组合数据集（QASC 和 Bamboogle）中的事实，研究了 LLaMA 2 和 Flan-T5 两种架构的传递性推理能力。控制了可能影响模型表现的三种线索：(a) 测试输入各部分间的词语/短语重叠；(b) 预训练或微调期间模型固有的知识；(c) 命名实体。结果表明，两者均依赖 (a)，但 Flan-T5 在 (b) 和 (c) 实验中表现出更强的鲁棒性，方差显著低于 LLaMA 2。这提示模型可能通过在相关数据集上的微调获得对传递性的理解，该假设有待后续研究验证。
原文摘要 · Abstract (English)
Evaluating Large Language Models (LLMs) on reasoning benchmarks demonstrates their ability to solve compositional questions. However, little is known of whether these models engage in genuine logical reasoning or simply rely on implicit cues to generate answers. In this paper, we investigate the transitive reasoning capabilities of two distinct LLM architectures, LLaMA 2 and Flan-T5, by manipulating facts within two compositional datasets: QASC and Bamboogle. We controlled for potential cues that might influence the models' performance, including (a) word/phrase overlaps across sections of test input; (b) models' inherent knowledge during pre-training or fine-tuning; and (c) Named Entities. Our findings reveal that while both models leverage (a), Flan-T5 shows more resilience to experiments (b and c), having less variance than LLaMA 2. This suggests that models may develop an understanding of transitivity through fine-tuning on knowingly relevant datasets, a hypothesis we leave to future work.
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。