LLMPhy: 结合大语言模型和物理引擎的参数可识别物理推理
LLMPhy: Parameter-Identifiable Physical Reasoning Combining Large Language Models and Physics Engines
AI总结：
LLMPhy通过整合大语言模型与物理引擎，解决复杂物理推理中参数识别问题，提出数字孪生构建方法，引入新数据集验证性能，实现更准确的参数恢复与稳定收敛。
摘要：
大多数基于学习的复杂物理推理方法回避了参数识别（如质量、摩擦力）这一关键问题，尽管这对现实应用如避障和机器人操作至关重要。本文提出LLMPhy，一种黑盒优化框架，将大语言模型（LLMs）与物理模拟器结合进行物理推理。LLMPhy的核心思想是将嵌入在LLMs中的教科书物理知识与现代物理引擎中的世界模型相结合，通过隐含参数估计构建输入场景的数字孪生。具体而言，LLMPhy将数字孪生构建分解为两个子问题：（i）连续问题，即估计物理参数；（ii）离散问题，即估计场景布局。对于每个子问题，LLMPhy迭代地提示LLM生成编码参数估计的计算机程序，将其在物理引擎中执行以重建场景，并利用由此产生的重建误差作为反馈来改进LLM的预测。由于现有物理推理基准很少考虑参数可识别性，我们引入了三个新数据集来评估零样本设置下的物理推理。我们的结果表明，LLMPhy在我们的任务中实现了最先进的性能，比现有黑盒方法更准确地恢复物理参数，并且收敛更可靠。详见LLMPhy项目页面：https://www.merl.com/research/highlights/LLMPhy
英文摘要：
Most learning-based approaches to complex physical reasoning sidestep the crucial problem of parameter identification (e.g., mass, friction) that governs scene dynamics, despite its importance in real-world applications such as collision avoidance and robotic manipulation. In this paper, we present LLMPhy, a black-box optimization framework that integrates large language models (LLMs) with physics simulators for physical reasoning. The core insight of LLMPhy is to bridge the textbook physical knowledge embedded in LLMs with the world models implemented in modern physics engines, enabling the construction of digital twins of input scenes via latent parameter estimation. Specifically, LLMPhy decomposes digital twin construction into two subproblems: (i) a continuous problem of estimating physical parameters and (ii) a discrete problem of estimating scene layout. For each subproblem, LLMPhy iteratively prompts the LLM to generate computer programs encoding parameter estimates, executes them in the physics engine to reconstruct the scene, and uses the resulting reconstruction error as feedback to refine the LLM's predictions. As existing physical reasoning benchmarks rarely account for parameter identifiability, we introduce three new datasets designed to evaluate physical reasoning in zero-shot settings. Our results show that LLMPhy achieves state-of-the-art performance on our tasks, recovers physical parameters more accurately, and converges more reliably than prior black-box methods. See the LLMPhy project page for details: https://www.merl.com/research/highlights/LLMPhy
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。
