Teaching Physical Awareness to LLMs through Sounds

Weiguo Wang, Andy Nie, Wenrui Zhou, Yi Kai, Chengchen Hu
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:63952-63971, 2025.

Abstract

Large Language Models (LLMs) have shown remarkable capabilities in text and multimodal processing, yet they fundamentally lack physical awareness–understanding of real-world physical phenomena. In this work, we present ACORN, a framework that teaches LLMs physical awareness through sound, focusing on fundamental physical phenomena like the Doppler effect, multipath effect, and spatial relationships. To overcome data scarcity, ACORN introduce a physics-based simulator combining real-world sound sources with controlled physical channels to generate diverse training data. Using this simulator, we build AQA-PHY, a comprehensive Audio Question-Answer dataset, and propose an audio encoder that processes both magnitude and phase information. By connecting our audio encoder to state-of-the-art LLMs, we demonstrate reasonable results in both simulated and real-world tasks, such as line-of-sight detection, Doppler effect estimation, and Direction-of-Arrival estimation, paving the way for enabling LLMs to understand physical world.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-wang25ca, title = {Teaching Physical Awareness to {LLM}s through Sounds}, author = {Wang, Weiguo and Nie, Andy and Zhou, Wenrui and Kai, Yi and Hu, Chengchen}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {63952--63971}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/wang25ca/wang25ca.pdf}, url = {https://proceedings.mlr.press/v267/wang25ca.html}, abstract = {Large Language Models (LLMs) have shown remarkable capabilities in text and multimodal processing, yet they fundamentally lack physical awareness–understanding of real-world physical phenomena. In this work, we present ACORN, a framework that teaches LLMs physical awareness through sound, focusing on fundamental physical phenomena like the Doppler effect, multipath effect, and spatial relationships. To overcome data scarcity, ACORN introduce a physics-based simulator combining real-world sound sources with controlled physical channels to generate diverse training data. Using this simulator, we build AQA-PHY, a comprehensive Audio Question-Answer dataset, and propose an audio encoder that processes both magnitude and phase information. By connecting our audio encoder to state-of-the-art LLMs, we demonstrate reasonable results in both simulated and real-world tasks, such as line-of-sight detection, Doppler effect estimation, and Direction-of-Arrival estimation, paving the way for enabling LLMs to understand physical world.} }
Endnote
%0 Conference Paper %T Teaching Physical Awareness to LLMs through Sounds %A Weiguo Wang %A Andy Nie %A Wenrui Zhou %A Yi Kai %A Chengchen Hu %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-wang25ca %I PMLR %P 63952--63971 %U https://proceedings.mlr.press/v267/wang25ca.html %V 267 %X Large Language Models (LLMs) have shown remarkable capabilities in text and multimodal processing, yet they fundamentally lack physical awareness–understanding of real-world physical phenomena. In this work, we present ACORN, a framework that teaches LLMs physical awareness through sound, focusing on fundamental physical phenomena like the Doppler effect, multipath effect, and spatial relationships. To overcome data scarcity, ACORN introduce a physics-based simulator combining real-world sound sources with controlled physical channels to generate diverse training data. Using this simulator, we build AQA-PHY, a comprehensive Audio Question-Answer dataset, and propose an audio encoder that processes both magnitude and phase information. By connecting our audio encoder to state-of-the-art LLMs, we demonstrate reasonable results in both simulated and real-world tasks, such as line-of-sight detection, Doppler effect estimation, and Direction-of-Arrival estimation, paving the way for enabling LLMs to understand physical world.
APA
Wang, W., Nie, A., Zhou, W., Kai, Y. & Hu, C.. (2025). Teaching Physical Awareness to LLMs through Sounds. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:63952-63971 Available from https://proceedings.mlr.press/v267/wang25ca.html.

Related Material