利用大型语言模型结合思维链与提示工程进行交通事故严重程度分析与推断
Leveraging Large Language Models with Chain-of-Thought and Prompt Engineering for Traffic Crash Severity Analysis and Inference
AI总结：
本文将事故严重程度推断建模为分类任务，利用模板叙述、CoT 与领域提示工程评估 GPT-3.5-turbo 及 LLaMA3 模型，证明 LLaMA3-70B 与 CoT 提示可显著提升推理性能。
摘要：
本研究借助大型语言模型（LLM）的能力，探索使用三个最先进的 LLM，即 GPT-3.5-turbo、LLaMA3-8B 和 LlaMA3-70B，来进行事故严重程度推断，并将其构建为一项分类任务。我们使用预先构建且融入领域知识的模板，从原始交通事故表格数据生成文本叙述。此外，我们引入思维链（CoT）推理，引导 LLM 分析事故原因并进而推断严重程度。本研究还考察了专为事故严重程度推断设计的提示工程所产生的影响。LLM 被要求执行事故严重程度推断，以：（1）评估模型在事故严重程度分析方面的能力；（2）评估 CoT 和基于领域知识的提示工程的有效性；（3）考察在 CoT 框架下的推理能力。结果表明，LLaMA3-70B 始终优于其他模型，尤其是在零样本设置下。CoT 和提示工程技术显著提升了性能，改善了逻辑推理并解决了对齐问题。值得注意的是，CoT 为理解 LLM 的推理过程提供了宝贵洞见，释放了它们在严重程度分析和推断中考虑多种因素的能力，例如环境条件、驾驶员行为和车辆特征。
英文摘要：
Harnessing the power of Large Language Models (LLMs), this study explores the use of three state-of-the-art LLMs, specifically GPT-3.5-turbo, LLaMA3-8B, and LLaMA3-70B, for crash severity inference, framing it as a classification task. We generate textual narratives from original traffic crash tabular data using a pre-built template infused with domain knowledge. Additionally, we incorporated Chain-of-Thought (CoT) reasoning to guide the LLMs in analyzing the crash causes and then inferring the severity. This study also examine the impact of prompt engineering specifically designed for crash severity inference. The LLMs were tasked with crash severity inference to: (1) evaluate the models' capabilities in crash severity analysis, (2) assess the effectiveness of CoT and domain-informed prompt engineering, and (3) examine the reasoning abilities with the CoT framework. Our results showed that LLaMA3-70B consistently outperformed the other models, particularly in zero-shot settings. The CoT and Prompt Engineering techniques significantly enhanced performance, improving logical reasoning and addressing alignment issues. Notably, the CoT offers valuable insights into LLMs' reasoning processes, unleashing their capacity to consider diverse factors such as environmental conditions, driver behavior, and vehicle characteristics in severity analysis and inference.
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。
