Achieving Gold-Medal-Level Olympiad Reasoning via Simple and Unified Scaling

TL;DR

This paper presents a unified scaling approach to transform reasoning models into olympiad-level problem solvers, achieving gold-medal performance on IMO and IPhO problems.

Contribution

It introduces a simple, scalable recipe combining curriculum fine-tuning and reinforcement learning to enhance reasoning models for complex scientific and mathematical problems.

Findings

Achieved gold-medal-level performance on IMO 2025 and IPhO 2024/2025.

Supported stable reasoning on problems with over 100K token trajectories.

Demonstrated strong generalization to scientific reasoning beyond math and physics.

Abstract

Recent progress in reasoning models has substantially advanced long-horizon mathematical and scientific problem solving, with several systems now reaching gold-medal-level performance on International Mathematical Olympiad (IMO) and International Physics Olympiad (IPhO) problems. In this paper, we introduce a simple and unified recipe for converting a post-trained reasoning backbone into a rigorous olympiad-level solver. The recipe first uses a reverse-perplexity curriculum for SFT to instill rigorous proof-search and self-checking behaviors, then scales these behaviors through a two-stage RL pipeline that progresses from RL with verifiable rewards to more delicate proof-level RL, and finally boosts solving performance with test-time scaling. Applying this recipe, we train a 30B-A3B backbone with SFT on around 340K sub-8K-token trajectories followed by 200 RL steps. The resulting model, SU-01, supports stable reasoning on difficult problems with trajectories exceeding 100K tokens, while achieving gold-medal-level performance on mathematical and physical olympiad competitions, including IMO 2025/USAMO 2026 and IPhO 2024/2025. It also demonstrates strong generalization of scientific reasoning to domains beyond mathematics and physics.