PhysElite: How Far Are LLMs from Solving Olympiad-Level Physics Problems?
cs.AI, cs.MM, math-ph, math.MP
Submitted: 2026-08-25
Updated: 2026-09-25
License: http://creativecommons.org/licenses/by/4.0/
The gist: Understanding how (multimodal) large language models perform on physics problems requires benchmarks that reflect the difficulty and breadth of expert-level physical reasoning.
Terminology
Abstract
Understanding how (multimodal) large language models perform on physics problems requires benchmarks that reflect the difficulty and breadth of expert-level physical reasoning. Existing physics benchmarks remain limited in the following two important ways: (1) short of high-difficulty datasets, and (2) lack of comprehensive coverage of visual forms, knowledge points, and step-by-step solution processes. As a result, model performance on current datasets may not be fully representative of their ability to solve complex physics problems. To address these issues, we present PhysElite, a large-scale bilingual multimodal benchmark for Olympiad-level physics reasoning. PhysElite contains 11,586 Olympiad-tier problems. For each problem, we provide corresponding visual diagrams, step-by-step bilingual Chinese-English solution derivations, and the final answer. We benchmark 18 open-source and closed-source MLLMs, and find that even the strongest model reaches only 33.7% answer accuracy. We additionally conduct step-level process evaluation to diagnose where models fail in the reasoning chain. Our datasets are released at https://huggingface.co/datasets/physelite/PhysElite.
Sources
- HiPhO: How Far Are (M)LLMs from Humans in the Latest High School Physics Olympiad Benchmark?
- PhysUniBench: A Multi-Modal Physics Reasoning Benchmark at Undergraduate Level
- UGPhysics: A Comprehensive Benchmark for Undergraduate Physics Reasoning with Large Language Models
- P1: Mastering Physics Olympiads with Reinforcement Learning
- Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering
- PhyX: Does Your Model Have the "Wits" for Physical Reasoning?
- Benchmarking Foundation Models with Retrieval-Augmented Generation in Olympic-Level Physics Problem Solving
- Kimi K2: Open Agentic Intelligence
- Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
- Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond
- GPT-4o System Card
- Qwen3-VL Technical Report
- DeepSeek-V3 Technical Report
- Qwen2.5-VL Technical Report
- The Llama 3 Herd of Models
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems
- VSAL: A Vision Solver with Adaptive Layouts for Graph Property Detection