LLM agents can partially mimic human driving behaviors and decision ...
5 days ago · Our evaluation and data attribution framework guides the development of interpretable, human-like LLM agents for smoother traffic integration.
LanguageMPC: Large Language Models as Decision Makers for …
Oct 4, 2023 · This paper presents an initial step toward leveraging LLMs as effective decision-makers for intricate AD scenarios in terms of safety, efficiency, generalizability, and interoperability.
Driving with LLMs: Fusing Object-Level Vector Modality for ... - GitHub
Jan 28, 2024 · The LLM-Driver utilises object-level vector input from our driving simulator to predict explanable actions using pretrained Language Models, providing a robust and interpretable solution …
Integrated LLM Based Reasoning for safe and Efficient Decision …
Jan 9, 2025 · Large Language Models (LLMs) represent the best hope for augmenting Advanced Driver Assistance Systems (ADAS) with powerful reasoning that is nonetheless fully
We devise cognitive pathways to en-able comprehensive reasoning with LLMs, and develop algorithms for translating LLM decisions into actionable driving commands.
Making Large Language Models Better Planners with Reasoning-Decision …
Inspired by the knowledge-driven nature of human driving, recent approaches explore the potential of large language models (LLMs) to improve understanding and decision-making in traffic scenarios.
Tackle Complex LLM Decision-Making with Language Agent Tree …
Aug 26, 2024 · Large Language Models (LLMs) have demonstrated exceptional abilities in performing natural language tasks that involve complex reasoning. As a result, these models have evolved to …
LEAD: LLM-enhanced deep reinforcement learning for stable decision ...
Dec 7, 2025 · To build a highly reliable, low-cost, and low-latency high-level intelligent LLM-based agent, we selected several mainstream LLMs provided by different vendors and deployed them to handle …
Dual-process theory and decision-making in large language models
Nov 14, 2025 · In this Review, we examine LLM outputs through the lens of dual-process theory and against the backdrop of human decision-making.
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.