Chain-of-Planned-Behaviour 工作流引发 LLM 的少样本移动性生成
Chain-of-Planned-Behaviour Workflow Elicits Few-Shot Mobility Generation in LLMs
AI总结：
受计划行为理论启发提出 CoPB 工作流，利用认知结构增强 LLM 移动意图推理，将错误率从 57.8% 降至 19.4%，并结合重力模型减少 47.0% 的 token 成本，同时支持自动生成标签微调开源模型以降低使用成本。
摘要：
大型语言模型（LLMs）强大的推理能力为许多领域带来了革命性的变化，但其在人类行为生成方面的性能尚未得到广泛探索。这一差距的出现，可能是因为支配行为意图的内部过程不能仅仅通过抽象推理来解释。相反，它们还受到包括社会规范和个人偏好在内的众多因素的影响。受计划行为理论（TPB）的启发，我们开发了一个名为 Chain-of-Planned Behaviour (CoPB) 的 LLM 工作流，用于移动行为生成，该工作流反映了人类活动的重要时空动态。通过利用 TPB 中的态度、主观规范和感知行为控制等认知结构，CoPB 显著增强了 LLM 推理下一步移动意图的能力。具体而言，CoPB 将移动意图生成的错误率从 57.8% 大幅降低至 19.4%。为了提高所提出的 CoPB 工作流的可扩展性，我们进一步探索了 LLM 与机制模型之间的协同作用。我们发现，诸如重力模型之类的机制移动模型可以有效地将移动意图映射到物理移动行为。整合 CoPB 与重力模型的策略可以减少 47.0% 的 token 成本并实现相当的性能。此外，所提出的 CoPB 工作流可以促进 GPT-3.5-turbo 自动为移动行为推理生成高质量标签。我们展示了此类标签可用于微调小规模、开源的 LLaMA 3 8B 模型，在不牺牲生成行为质量的情况下显著降低使用成本。
英文摘要：
The powerful reasoning capabilities of large language models (LLMs) have brought revolutionary changes to many fields, but their performance in human behaviour generation has not yet been extensively explored. This gap likely emerges because the internal processes governing behavioral intentions cannot be solely explained by abstract reasoning. Instead, they are also influenced by a multitude of factors, including social norms and personal preference. Inspired by the Theory of Planned Behaviour (TPB), we develop a LLM workflow named Chain-of-Planned Behaviour (CoPB) for mobility behaviour generation, which reflects the important spatio-temporal dynamics of human activities. Through exploiting the cognitive structures of attitude, subjective norms, and perceived behaviour control in TPB, CoPB significantly enhance the ability of LLMs to reason the intention of next movement. Specifically, CoPB substantially reduces the error rate of mobility intention generation from 57.8% to 19.4%. To improve the scalability of the proposed CoPB workflow, we further explore the synergy between LLMs and mechanistic models. We find mechanistic mobility models, such as gravity model, can effectively map mobility intentions to physical mobility behaviours. The strategy of integrating CoPB with gravity model can reduce the token cost by 97.7% and achieve better performance simultaneously. Besides, the proposed CoPB workflow can facilitate GPT-4-turbo to automatically generate high quality labels for mobility behavior reasoning. We show such labels can be leveraged to fine-tune the smaller-scale, open source LLaMA 3-8B, which significantly reduces usage costs without sacrificing the quality of the generated behaviours.
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。
