SOURCE-LINKED INTELLIGENCE

An Open Recipe for IMO Gold: Training Nemotron for Olympiad Mathematics

arXiv · AI, language, vision and robotics · article · Sep 9, 2026 · UTC

We study how model post-training and test-time inference design affect natural-language proof generation for hard olympiad mathematics. Starting from Nemotron 3 Ultra, we train two specialist checkpoints using supervised fine-tuning and reinforcement learning, and evaluate checkpoint choice, verification, and refinement. Based on these findings, we present an open-model test-time-compute pipeline. The system operates entirely in natural language, with no formal prover, external tools, or internet access. Three Nemotron 3 Ultra checkpoints - the general-availability model and two post-trained s

Read original source ↗ Open in workspace

recordType

paper

region

Global