Physics Supernova: AI Agent Matches Elite Gold Medalists at IPhO 2025

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Qiu, Jiahao, Shi, Jingzhe, Juan, Xinzhe, Zhao, Zelin, Geng, Jiayi, Liu, Shilong, Wang, Hongru, Wu, Sanfeng, Wang, Mengdi
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866915474488426496
author Qiu, Jiahao
Shi, Jingzhe
Juan, Xinzhe
Zhao, Zelin
Geng, Jiayi
Liu, Shilong
Wang, Hongru
Wu, Sanfeng
Wang, Mengdi
author_facet Qiu, Jiahao
Shi, Jingzhe
Juan, Xinzhe
Zhao, Zelin
Geng, Jiayi
Liu, Shilong
Wang, Hongru
Wu, Sanfeng
Wang, Mengdi
contents Physics provides fundamental laws that describe and predict the natural world. AI systems aspiring toward more general, real-world intelligence must therefore demonstrate strong physics problem-solving abilities: to formulate and apply physical laws for explaining and predicting physical processes. The International Physics Olympiad (IPhO)--the world's most prestigious physics competition--offers a rigorous benchmark for this purpose. We introduce Physics Supernova, an AI agent system with superior physics problem-solving abilities that match elite IPhO gold medalists. In IPhO 2025 theory problems, Physics Supernova attains 23.5/30 points, ranking 14th of 406 contestants and surpassing the median performance of human gold medalists. We extensively analyzed Physics Supernova's capabilities and flexibility across diverse physics tasks. These results show that principled tool integration within agent systems can deliver competitive improvements in solving challenging science problems. The codes are available at https://github.com/CharlesQ9/Physics-Supernova.
format Preprint
id arxiv_https___arxiv_org_abs_2509_01659
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Physics Supernova: AI Agent Matches Elite Gold Medalists at IPhO 2025
Qiu, Jiahao
Shi, Jingzhe
Juan, Xinzhe
Zhao, Zelin
Geng, Jiayi
Liu, Shilong
Wang, Hongru
Wu, Sanfeng
Wang, Mengdi
Artificial Intelligence
Physics provides fundamental laws that describe and predict the natural world. AI systems aspiring toward more general, real-world intelligence must therefore demonstrate strong physics problem-solving abilities: to formulate and apply physical laws for explaining and predicting physical processes. The International Physics Olympiad (IPhO)--the world's most prestigious physics competition--offers a rigorous benchmark for this purpose. We introduce Physics Supernova, an AI agent system with superior physics problem-solving abilities that match elite IPhO gold medalists. In IPhO 2025 theory problems, Physics Supernova attains 23.5/30 points, ranking 14th of 406 contestants and surpassing the median performance of human gold medalists. We extensively analyzed Physics Supernova's capabilities and flexibility across diverse physics tasks. These results show that principled tool integration within agent systems can deliver competitive improvements in solving challenging science problems. The codes are available at https://github.com/CharlesQ9/Physics-Supernova.
title Physics Supernova: AI Agent Matches Elite Gold Medalists at IPhO 2025
topic Artificial Intelligence
url https://arxiv.org/abs/2509.01659