Making Large Language Models Better Planners with Reasoning-Decision Alignment

• URL: http://arxiv.org/abs/2408.13890v1
• Date: Sun, 25 Aug 2024 16:43:47 GMT
• Title: Making Large Language Models Better Planners with Reasoning-Decision Alignment
• Authors: Zhijian Huang, Tao Tang, Shaoxiang Chen, Sihao Lin, Zequn Jie, Lin Ma, Guangrun Wang, Xiaodan Liang,
• Abstract summary: We motivate an end-to-end decision-making model based on multimodality-augmented LLM. We propose a reasoning-decision alignment constraint between the paired CoTs and planning results. We dub our proposed large language planners with reasoning-decision alignment as RDA-Driver.
• Score: 70.5381163219608
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Data-driven approaches for autonomous driving (AD) have been widely adopted in the past decade but are confronted with dataset bias and uninterpretability. 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. They find that the pretrain-finetune paradigm of LLMs on downstream data with the Chain-of-Thought (CoT) reasoning process can enhance explainability and scene understanding. However, such a popular strategy proves to suffer from the notorious problems of misalignment between the crafted CoTs against the consequent decision-making, which remains untouched by previous LLM-based AD methods. To address this problem, we motivate an end-to-end decision-making model based on multimodality-augmented LLM, which simultaneously executes CoT reasoning and carries out planning results. Furthermore, we propose a reasoning-decision alignment constraint between the paired CoTs and planning results, imposing the correspondence between reasoning and decision-making. Moreover, we redesign the CoTs to enable the model to comprehend complex scenarios and enhance decision-making performance. We dub our proposed large language planners with reasoning-decision alignment as RDA-Driver. Experimental evaluations on the nuScenes and DriveLM-nuScenes benchmarks demonstrate the effectiveness of our RDA-Driver in enhancing the performance of end-to-end AD systems. Specifically, our RDA-Driver achieves state-of-the-art planning performance on the nuScenes dataset with 0.80 L2 error and 0.32 collision rate, and also achieves leading results on challenging DriveLM-nuScenes benchmarks with 0.82 L2 error and 0.38 collision rate.

Related papers

• Unlocking the Capabilities of Thought: A Reasoning Boundary Framework to Quantify and Optimize Chain-of-Thought [61.588465852846646]
  Chain-of-Thought (CoT) reasoning has emerged as a promising approach for enhancing the performance of large language models (LLMs) In this work, we introduce a novel reasoning boundary framework (RBF) to address these challenges.
  arXiv  Detail & Related papers  (2024-10-08T05:26:28Z)
• Deliberate Reasoning for LLMs as Structure-aware Planning with Accurate World Model [14.480267340831542]
  We propose Structure-aware Planning with Accurate World Model (SWAP) for large language models (LLMs) SWAP incorporates structural information to guide the reasoning process via a world model and provides a soft verification mechanism over the steps. We evaluate SWAP across diverse reasoning-intensive benchmarks including math reasoning, logical reasoning, and coding tasks.
  arXiv  Detail & Related papers  (2024-10-04T04:23:36Z)
• The Role of Deductive and Inductive Reasoning in Large Language Models [35.43513487137371]
  Large Language Models (LLMs) have achieved substantial progress in artificial intelligence, particularly in reasoning tasks. We propose the Deductive and InDuctive(DID) method, which enhances LLM reasoning by dynamically integrating both deductive and inductive reasoning. Our findings suggest that DID provides a more robust and cognitively aligned framework for reasoning in LLMs.
  arXiv  Detail & Related papers  (2024-10-03T18:30:47Z)
• Towards Interactive and Learnable Cooperative Driving Automation: a Large Language Model-Driven Decision-Making Framework [79.088116316919]
  Connected Autonomous Vehicles (CAVs) have begun to open road testing around the world, but their safety and efficiency performance in complex scenarios is still not satisfactory. This paper proposes CoDrivingLLM, an interactive and learnable LLM-driven cooperative driving framework.
  arXiv  Detail & Related papers  (2024-09-19T14:36:00Z)
• Bridging and Modeling Correlations in Pairwise Data for Direct Preference Optimization [75.1240295759264]
  We propose an effective framework for Bridging and Modeling Correlations in pairwise data, named BMC. We increase the consistency and informativeness of the pairwise preference signals through targeted modifications. We identify that DPO alone is insufficient to model these correlations and capture nuanced variations.
  arXiv  Detail & Related papers  (2024-08-14T11:29:47Z)
• Tokenize the World into Object-level Knowledge to Address Long-tail Events in Autonomous Driving [43.156632952193966]
  Traditional end-to-end driving models suffer from long-tail events due to rare or unseen inputs within their training distributions. We propose TOKEN, a novel Multi-Modal Large Language Model (MM-LLM) that tokenizes the world into object-level knowledge. ToKEN effectively alleviates data scarcity and inefficient tokenization by leveraging a traditional end-to-end driving model.
  arXiv  Detail & Related papers  (2024-07-01T04:34:50Z)
• P-TA: Using Proximal Policy Optimization to Enhance Tabular Data Augmentation via Large Language Models [15.969452637480167]
  We propose using proximal policy optimization (PPO) to apply Generative Adversarial Networks (GANs) PPO leads to an approximately 4% improvement in the accuracy of models trained on synthetically generated data over state-of-the-art datasets.
  arXiv  Detail & Related papers  (2024-06-17T10:22:00Z)
• Modeling Boundedly Rational Agents with Latent Inference Budgets [56.24971011281947]
  We introduce a latent inference budget model (L-IBM) that models agents' computational constraints explicitly. L-IBMs make it possible to learn agent models using data from diverse populations of suboptimal actors. We show that L-IBMs match or outperform Boltzmann models of decision-making under uncertainty.
  arXiv  Detail & Related papers  (2023-12-07T03:55:51Z)
• Reinforcement Learning with a Terminator [80.34572413850186]
  We learn the parameters of the TerMDP and leverage the structure of the estimation problem to provide state-wise confidence bounds. We use these to construct a provably-efficient algorithm, which accounts for termination, and bound its regret.
  arXiv  Detail & Related papers  (2022-05-30T18:40:28Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.