LLM在决策中表现出顽固偏见，影响人机协作效果。
Rigidity in LLM Bandits with Implications for Human-AI Dyads
- 用双臂老虎机测试模型决策行为，发现位置偏好被放大为固定选择。
- 在不对称奖励下，模型僵化执行策略且表现低于最优基准。
- 揭示其学习率低、温度极高，适合研究人机互动中的认知偏差。
我们检验了大语言模型是否具有稳健的决策偏见。将模型置于双臂老虎机任务中，每种条件运行20000次试验，涵盖四种解码配置。在对称奖励下，模型将位置顺序放大为顽固的单臂偏好；在非对称奖励下，它们表现出僵化的利用策略，表现低于神谕（oracle），且很少重新验证。这些模式在温度与top-p的调整下保持一致，而top-k维持默认值，表明这些定性行为对实际使用者可调参数具有鲁棒性。关键的是，通过层次化Rescorla-Wagner-softmax建模发现，其底层策略为低学习率与极高逆温度，共同解释了噪声到偏见的放大及僵化利用现象。这表明最小化老虎机任务是探测模型决策倾向的可行工具，并为人类-人工智能互动中的偏见形成提供了假设基础。
原文摘要 · Abstract (English)
We test whether LLMs show robust decision biases. Treating models as participants in two-arm bandits, we ran 20000 trials per condition across four decoding configurations. Under symmetric rewards, models amplified positional order into stubborn one-arm policies. Under asymmetric rewards, they exploited rigidly yet underperformed an oracle and rarely re-checked. The observed patterns were consistent across manipulations of temperature and top-p, with top-k held at the provider default, indicating that the qualitative behaviours are robust to the two decoding knobs typically available to practitioners. Crucially, moving beyond descriptive metrics to computational modelling, a hierarchical Rescorla-Wagner-softmax fit revealed the underlying strategies: low learning rates and very high inverse temperatures, which together explain both noise-to-bias amplification and rigid exploitation. These results position minimal bandits as a tractable probe of LLM decision tendencies and motivate hypotheses about how such biases could shape human-AI interaction.
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。