大型语言模型是否具有人类般的偏见？在先验知识、无关上下文和不同计算预算下的因果推理
Do LLMs Share Human-Like Biases? Causal Reasoning Under Prior Knowledge, Irrelevant Context, and Varying Compute Budgets
AI总结：
本文通过对比20余种LLM与人类基准，在11个由碰撞结构形式化的因果判断任务中发现，小型可解释模型能有效压缩LLM的因果判断，多数LLM表现出比人类更规则化的推理策略，但未表现出人类典型的碰撞偏见。
摘要：
大型语言模型（LLMs）正越来越多地应用于需要因果推理的领域，但尚不清楚其判断反映的是规范性因果计算、人类般的快捷方式还是脆弱的模式匹配。我们通过11个由碰撞结构（C₁→E←C₂）形式化的因果判断任务，对20余种LLM与匹配的人类基准进行对比测试。我们发现，一个小型可解释模型能够有效压缩LLM的因果判断，且大多数LLM表现出比人类更规则化的推理策略，而人类似乎在概率判断中考虑了未提及的潜在因素。此外，大多数LLM并未表现出人类典型的碰撞偏见，如弱解释消失和马尔可夫违反。我们进一步测试了LLM在（i）语义抽象和（ii）提示过载（注入无关文本）下的因果判断鲁棒性，并发现对于许多LLM而言，链式推理（CoT）能提高鲁棒性。总体而言，这种差异表明，当已知偏见不希望时，LLM可以补充人类，但其规则化推理在存在内在不确定性时可能会失效——这突显了需要表征LLM推理策略以实现安全、有效部署的必要性。
英文摘要：
Large language models (LLMs) are increasingly used in domains where causal reasoning matters, yet it remains unclear whether their judgments reflect normative causal computation, human-like shortcuts, or brittle pattern matching. We benchmark 20+ LLMs against a matched human baseline on 11 causal judgment tasks formalized by a collider structure ($C_1 \rightarrow E \leftarrow C_2$). We find that a small interpretable model compresses LLMs' causal judgments well and that most LLMs exhibit more rule-like reasoning strategies than humans who seem to account for unmentioned latent factors in their probability judgments. Furthermore, most LLMs do not mirror the characteristic human collider biases of weak explaining away and Markov violations. We probe LLMs' causal judgment robustness under (i) semantic abstraction and (ii) prompt overloading (injecting irrelevant text), and find that chain-of-thought (CoT) increases robustness for many LLMs. Together, this divergence suggests LLMs can complement humans when known biases are undesirable, but their rule-like reasoning may break down when uncertainty is intrinsic - highlighting the need to characterize LLM reasoning strategies for safe, effective deployment.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
