PhysicsMinions: Winning Gold Medals in the Latest Physics Olympiads with a Coevolutionary Multimodal Multi-Agent System
- URL: http://arxiv.org/abs/2509.24855v1
- Date: Mon, 29 Sep 2025 14:40:53 GMT
- Title: PhysicsMinions: Winning Gold Medals in the Latest Physics Olympiads with a Coevolutionary Multimodal Multi-Agent System
- Authors: Fangchen Yu, Junchi Yao, Ziyi Wang, Haiyuan Wan, Youling Huang, Bo Zhang, Shuyue Hu, Dongzhan Zhou, Ning Ding, Ganqu Cui, Lei Bai, Wanli Ouyang, Peng Ye,
- Abstract summary: Physics is central to understanding and shaping the real world, and the ability to solve physics problems is a key indicator of real-world physical intelligence.<n>Existing approaches are predominantly single-model based, and open-source MLLMs rarely reach gold-medal-level performance.<n>We propose PhysicsMinions, a coevolutionary multi-agent system for Physics Olympiad.<n>Its architecture features three synergistic studios: a Visual Studio to interpret diagrams, a Logic Studio to formulate solutions, and a Review Studio to perform dual-stage verification.
- Score: 65.02248709992442
- License: http://creativecommons.org/licenses/by-nc-sa/4.0/
- Abstract: Physics is central to understanding and shaping the real world, and the ability to solve physics problems is a key indicator of real-world physical intelligence. Physics Olympiads, renowned as the crown of competitive physics, provide a rigorous testbed requiring complex reasoning and deep multimodal understanding, yet they remain largely underexplored in AI research. Existing approaches are predominantly single-model based, and open-source MLLMs rarely reach gold-medal-level performance. To address this gap, we propose PhysicsMinions, a coevolutionary multi-agent system for Physics Olympiad. Its architecture features three synergistic studios: a Visual Studio to interpret diagrams, a Logic Studio to formulate solutions, and a Review Studio to perform dual-stage verification. The system coevolves through an iterative refinement loop where feedback from the Review Studio continuously guides the Logic Studio, enabling the system to self-correct and converge towards the ground truth. Evaluated on the HiPhO benchmark spanning 7 latest physics Olympiads, PhysicsMinions delivers three major breakthroughs: (i) Strong generalization: it consistently improves both open-source and closed-source models of different sizes, delivering clear benefits over their single-model baselines; (ii) Historic breakthroughs: it elevates open-source models from only 1-2 to 6 gold medals across 7 Olympiads, achieving the first-ever open-source gold medal in the latest International Physics Olympiad (IPhO) under the average-score metric; and (iii) Scaling to human expert: it further advances the open-source Pass@32 score to 26.8/30 points on the latest IPhO, ranking 4th of 406 contestants and far surpassing the top single-model score of 22.7 (ranked 22nd). Generally, PhysicsMinions offers a generalizable framework for Olympiad-level problem solving, with the potential to extend across disciplines.
Related papers
- P1: Mastering Physics Olympiads with Reinforcement Learning [84.08897284032724]
We introduce P1, a family of open-source physics reasoning models trained entirely through reinforcement learning (RL)<n>P1-235B-A22B is the first open-source model with Gold-medal performance at the latest International Physics Olympiad (IPhO 2025), and wins 12 gold medals out of 13 international/regional physics competitions in 2024/2025.<n>P1-235B-A22B+PhysicsMinions achieves overall No.1 on IPhO 2025, and obtains the highest average score over the 13 physics competitions.
arXiv Detail & Related papers (2025-11-17T17:18:13Z) - HiPhO: How Far Are (M)LLMs from Humans in the Latest High School Physics Olympiad Benchmark? [53.76627321546095]
HiPhO is the first benchmark dedicated to high school physics Olympiads with human-aligned evaluation.<n>It compiles 13 latest Olympiad exams from 2024-2025, spanning both international and regional competitions.<n>We assign gold, silver, and bronze medals to models based on official medal thresholds, thereby enabling direct comparison between (M)LLMs and human contestants.
arXiv Detail & Related papers (2025-09-09T16:24:51Z) - Physics Supernova: AI Agent Matches Elite Gold Medalists at IPhO 2025 [55.8464246603186]
We introduce Physics Supernova, an AI system with superior physics problem-solving abilities.<n>Supernova attains 23.5/30 points, ranking 14th of 406 contestants and surpassing the median performance of human gold medalists.<n>These results show that principled tool integration within agent systems can deliver competitive improvements.
arXiv Detail & Related papers (2025-09-01T17:59:13Z) - PhysUniBench: An Undergraduate-Level Physics Reasoning Benchmark for Multimodal Models [69.73115077227969]
We present PhysUniBench, a large-scale benchmark designed to evaluate and improve the reasoning capabilities of large language models (MLLMs)<n>PhysUniBench consists of 3,304 physics questions spanning 8 major sub-disciplines of physics, each accompanied by one visual diagram.<n>The benchmark's construction involved a rigorous multi-stage process, including multiple roll-outs, expert-level evaluation, automated filtering of easily solved problems, and a nuanced difficulty grading system with five levels.
arXiv Detail & Related papers (2025-06-21T09:55:42Z) - OlympiadBench: A Challenging Benchmark for Promoting AGI with Olympiad-Level Bilingual Multimodal Scientific Problems [62.06169250463104]
We present OlympiadBench, an Olympiad-level bilingual multimodal scientific benchmark, featuring 8,476 problems from Olympiad-level mathematics and physics competitions.
The best-performing model, GPT-4V, attains an average score of 17.97% on OlympiadBench, with a mere 10.74% in physics.
Our analysis orienting GPT-4V points out prevalent issues with hallucinations, knowledge omissions, and logical fallacies.
arXiv Detail & Related papers (2024-02-21T18:49:26Z)
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.