# How Good Are Frontier Models at Physics? Expert Re-Grading Reveals Broken Evaluations and Near-Saturation of Leading Benchmarks

arXiv: https://arxiv.org/abs/2609.13009 · HTML: https://mlsanity.com/paper/2609.13009/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning.
- Builds on 13 of its 28 references that are in the map.
- First posted on arXiv 11 Sep 2026.

## Placed in

- LLM reasoning / [Spatial and embodied reasoning](https://mlsanity.com/lr/spatial-and-embodied-reasoning/)

## Builds on (most cited first, 13 of 13)

- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [SWE-bench: Can Language Models Resolve Real-World GitHub Issues?](https://mlsanity.com/paper/2310.06770/) (2023)
- [Humanity's Last Exam](https://mlsanity.com/paper/2501.14249/) (2025)
- [Pervasive Label Errors in Test Sets Destabilize Machine Learning Benchmarks](https://mlsanity.com/paper/2103.14749/) (2021)
- [Humans or LLMs as the Judge? A Study on Judgement Biases](https://mlsanity.com/paper/2402.10669/) (2024)
- [JudgeBench: A Benchmark for Evaluating LLM-based Judges](https://mlsanity.com/paper/2410.12784/) (2024)
- [What Will it Take to Fix Benchmarking in Natural Language Understanding?](https://mlsanity.com/paper/2104.02145/) (2021)
- [SWE-rebench: An Automated Pipeline for Task Collection and Decontaminated Evaluation of Software Engineering Agents](https://mlsanity.com/paper/2505.20411/) (2025)
- [UGPhysics: A Comprehensive Benchmark for Undergraduate Physics Reasoning with Large Language Models](https://mlsanity.com/paper/2502.00334/) (2025)
- [Semi-Autonomous Mathematics Discovery with Gemini: A Case Study on the ErdHos Problems](https://mlsanity.com/paper/2601.22401/) (2026)
- [Probing the Critical Point (CritPt) of AI Reasoning: a Frontier Physics Research Benchmark](https://mlsanity.com/paper/2509.26574/) (2025)
- [Solving an Open Problem in Theoretical Physics using AI-Assisted Discovery](https://mlsanity.com/paper/2603.04735/) (2026)
- [PRISM-Physics: Causal DAG-Based Process Evaluation for Physics Reasoning](https://mlsanity.com/paper/2510.03185/) (2025)

## Abstract (arXiv)

Low reported scores on leading physics benchmarks, including those featured in the Artificial Analysis Intelligence Index (2026), suggest that frontier language models still struggle with advanced physics, a demanding test of their scientific reasoning and quantitative problem-solving abilities. Yet this impression does not always align with domain experts' experiences using these models in their work. We revisit these reported findings by evaluating frontier models on six widely used physics benchmarks and auditing them with experts, focusing on text-only problems with verifiable final answers. For each subfield of physics, faculty and graduate researchers with relevant expertise carefully review problem statements, reference solutions, and model responses to distinguish genuine model errors from grader errors, incorrect reference solutions, and ambiguous or underspecified questions. Most audited cases initially evaluated as incorrect reflect these benchmarking issues rather than errors in the models' physics reasoning. We then ask experts to address these benchmarking issues by correcting erroneous reference solutions and repairing or excluding flawed questions. We find that GPT-5.6-Sol's measured mean@4 rises from 47.3% to 78.7% on HLE-Physics and from 61.0% to 87.2% on CMT-Benchmark, while its corrected pass@4 reaches 94.4% on the 54 retained CritPt challenges. Corrected scores are computed on the retained evaluation subsets following expert review. Scores on the audited subsets of UGPhysics, PRISM-Physics, and PHYBench also rise substantially after correction. These findings suggest that current benchmarks substantially understate frontier models' ability to solve well-posed physics problems. Near-saturation on these closed-ended tasks highlights the need for more demanding, expert-validated evaluations.