DRIV-EX:用于驾驶LLM的反事实解释
DRIV-EX: Counterfactual Explanations for Driving LLMs
AI总结:
本文提出DRIV-EX方法,通过反事实解释研究驾驶LLM的决策过程,利用梯度优化生成语义变化的场景描述,提升解释的可理解性和鲁棒性。
摘要:
大型语言模型(LLMs)越来越多地被用作自动驾驶中的推理引擎,但其决策过程仍不透明。我们提出通过反事实解释研究其决策过程,即确定场景描述中最小的语义变化以改变驾驶计划。我们引入了DRIV-EX方法,利用梯度优化连续嵌入来识别翻转模型决策所需的输入变化。关键在于避免非约束连续优化通常导致的不连贯文本,DRIV-EX将这些优化嵌入仅作为语义指南:它们用于偏指导受控解码过程重新生成原始场景描述。此方法有效将生成引导至反事实目标,同时保证语言流畅性、领域有效性及与原始输入的接近性,对可解释性至关重要。在LC-LLM规划器上评估文本转录的highD数据集时,DRIV-EX生成的有效、流畅的反事实解释比现有基线更可靠。它成功揭示了潜在偏见,并提供了改进基于LLM的驾驶代理鲁棒性的具体见解。代码可在
英文摘要:
Large language models (LLMs) are increasingly used as reasoning engines in autonomous driving, yet their decision-making remains opaque. We propose to study their decision process through counterfactual explanations, which identify the minimal semantic changes to a scene description required to alter a driving plan. We introduce DRIV-EX, a method that leverages gradient-based optimization on continuous embeddings to identify the input shifts required to flip the model's decision. Crucially, to avoid the incoherent text typical of unconstrained continuous optimization, DRIV-EX uses these optimized embeddings solely as a semantic guide: they are used to bias a controlled decoding process that re-generates the original scene description. This approach effectively steers the generation toward the counterfactual target while guaranteeing the linguistic fluency, domain validity, and proximity to the original input, essential for interpretability. Evaluated using the LC-LLM planner on a textual transcription of the highD dataset, DRIV-EX generates valid, fluent counterfactuals more reliably than existing baselines. It successfully exposes latent biases and provides concrete insights to improve the robustness of LLM-based driving agents. The code is available at "https://github.com/Amaia-CARDIEL/DRIV_EX" .
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。