AI Paper Atlas

Research map / Large language models / Agents, code and math

Sim2Reason: Solving Physics Olympiad via Reinforcement Learning on Physics Simulators

Mihir Prabhudesai, Aryan Satpathy, Yangmin Li, Zheyang Qin, Nikash Bhardwaj, Amir Zadeh, Chuan Li, Katerina Fragkiadaki, Deepak Pathak

ICML 2026 · Large language models › Agents, code and math · approach: Reasoning benchmarks and platforms

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
train with RL on numeric questions instead of reverse or symbolic (ablation)Qwen2.5-3B RL · IPhO Mechanicsaccuracy: 5.84% (reverse), 7.46% (symbolic) -> 13.15% (numeric)Improved
apply shortcut-solution filtering to generated QA (ablation)Qwen2.5-3B RL · IPhO Mechanicsaccuracy: 7.14% -> 13.15%Improved
use RL instead of SFT on 200k teacher-sampled QA (ablation)Qwen2.5-32B · IPhO Mechanicsaccuracy: 15.9% -> 25.2%Improved
SFT on teacher trajectories (ablation)Qwen2.5-32B · IPhO Mechanicsaccuracy: 19.8% -> 15.9%Hurt
train on simulator data instead of DAPO-17K math data (ablation)Qwen2.5-3B · IPhO Mechanicsaccuracy: 9.98 -> 13.15Improved
mix DAPO-17K with synthetic data vs synthetic only (ablation)Qwen2.5-3B · IPhO Mechanicsaccuracy: 13.15 -> 10.35Hurt
RL on synthetic simulator dataQwen2.5-32B · JEEBenchaccuracy: 34.38% -> 52.28%Improved
RL on synthetic simulator dataQwen2.5-32B · MATH 500accuracy: 78.4% -> 82.8%Improved
RL on synthetic simulator dataQwen3-30B · Synthetic Symbolicaccuracy: 8.8% -> 8.0%Hurt
RL on synthetic simulator dataQwen2.5-3B · IPhO Mechanicsaccuracy: 5.68% -> 13.15%Improved
RL on synthetic simulator dataQwen2.5-32B · PHYSICS Undergraduateaccuracy: 47.9% -> 48.4%Improved
RL on synthetic simulator dataQwen2.5-32B · PHYSICS Postgraduateaccuracy: 32.2% -> 37.8%Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy