将大模型模块迁移至自动驾驶轨迹生成，验证其有效性与适配需求。
Do LLM Modules Generalize? A Study on Motion Generation for Autonomous Driving
- 测试五类大模型模块在自动驾驶轨迹生成中的可迁移性。
- 在Waymo Sim Agents上实现竞争力表现，部分模块显著提升性能。
- 揭示哪些模块可直接迁移，哪些需适配，指导实际应用。
大语言模型（LLM）的最新进展不仅推动了自然语言处理，也启发其在结构相似领域（如自动驾驶轨迹生成）的应用。两者均涉及自回归序列建模、基于标记的表示和上下文感知决策，使得LLM组件的迁移成为自然且常见的实践。然而，尽管早期尝试表现出潜力，对哪些模块真正具备可迁移性的系统理解仍不足。本文在Waymo Sim Agents基准上，全面评估了五类关键LLM模块——分词器设计、位置编码、预训练范式、后训练策略及推理时计算——在自动驾驶轨迹生成中的表现。实验表明，经适当调整后，这些模块可显著提升生成性能。我们识别出可有效迁移的技术，分析其他失败原因，并讨论自动驾驶场景下的具体适配需求。方法在Sim Agents任务上取得竞争力结果。
原文摘要 · Abstract (English)
Recent breakthroughs in large language models (LLMs) have not only advanced natural language processing but also inspired their application in domains with structurally similar problems--most notably, autonomous driving motion generation. Both domains involve autoregressive sequence modeling, token-based representations, and context-aware decision making, making the transfer of LLM components a natural and increasingly common practice. However, despite promising early attempts, a systematic understanding of which LLM modules are truly transferable remains lacking. In this paper, we present a comprehensive evaluation of five key LLM modules--tokenizer design, positional embedding, pre-training paradigms, post-training strategies, and test-time computation--within the context of motion generation for autonomous driving. Through extensive experiments on the Waymo Sim Agents benchmark, we demonstrate that, when appropriately adapted, these modules can significantly improve performance for autonomous driving motion generation. In addition, we identify which techniques can be effectively transferred, analyze the potential reasons for the failure of others, and discuss the specific adaptations needed for autonomous driving scenarios. We evaluate our method on the Sim Agents task and achieve competitive results.
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。