HackerPapers

arXiv:2507.15855cs.AI

Winning Gold at IMO 2025 with a Model-Agnostic Verification-and-Refinement Pipeline

Yichen Huang, Lin F. Yang

Abstract

The International Mathematical Olympiad (IMO) is widely regarded as the world championship of high-school mathematics. IMO problems are renowned for their difficulty and novelty, demanding deep insight, creativity, and rigor. Although large language models perform well on many mathematical benchmarks, they often struggle with Olympiad-level problems. Using carefully designed prompts, we construct a model-agnostic, verification-and-refinement pipeline. We demonstrate its effectiveness on the recent IMO 2025, avoiding data contamination for models released before the competition. Equipped with any of the three leading models -- Gemini 2.5 Pro, Grok-4, or GPT-5 -- our pipeline correctly solved 5 out of the 6 problems ($\approx$85.7% accuracy). This is in sharp contrast to their baseline accuracies: 31.6% (Gemini 2.5 Pro), 21.4% (Grok-4), and 38.1% (GPT-5), obtained by selecting the best of 32 candidate solutions. The substantial improvement underscores that the path to advanced AI reasoning requires not only developing more powerful base models but also designing effective methodologies to harness their full potential for complex tasks.

arXiv comments
Journal reference
DOI
Subjects
cs.AIArtificial Intelligence

On Hacker News

Linked by 10 identifiable commenters in 10 comments across 9 threads, from Jul 2025 to Jan 2026. Found in 3 HN submissions from Jul 2025 to Jul 2025. Open submission discussions.

Comments linking this paper, per year.

Discussed in comments9 threads

  1. 2025: The Year in LLMs 1 comment

    • Except of course it's not true lol. Horses are smart critters, but they absolutely cannot do arithmetic no matter how much you train them. These things are not horses. How can anyone choose to remain so ignorant in the face of irrefutable evidence that they're wrong?…

      CamperBob2View on HNParent context

  2. Yann LeCun to depart Meta and launch AI startup focused on 'world models' 1 comment

    • a) no, gemini 2.5 was shown to "win" gold w/o tools. - https://arxiv.org/html/2507.15855v1 b) reductionism isn't worth our time. Planning works in the real world, today. (try any agentic tool like cc/codex/whatever). And if you're set on the purist view, there's mounting…

      NitpickLawyerView on HNParent context

  3. It's OpenAI's world, we're just living in it 1 comment

    • It's reasonable to question it, but there's a fun Chinese paper (https://arxiv.org/abs/2507.15855) where they attempt to create a system of prompts that can be used with commercial LLMs to solve hard maths problems. It turns out that they can use the same prompt system for all…

      impossibleforkView on HNParent context

  4. Implicit Actor Critic Coupling via a Supervised Learning Framework for RLVR 2 comments

    • I think you meant to link to Implicit Actor Critic Coupling via a Supervised Learning Framework for RLVR https://arxiv.org/abs/2509.02522 not Winning Gold at IMO 2025 with a Model-Agnostic Verification-and-Refinement Pipeline https://arxiv.org/abs/2507.15855

      yorwbaView on HNParent context

    • We've changed the top link to that from https://arxiv.org/abs/2507.15855. Thanks!

      dangView on HNParent context

  5. DeepMind and OpenAI win gold at ICPC 1 comment

    • A terse prompt and expecting a one-shot answer is really not how you'd get an LLM to solve complex problems. I don't know what Deepmind and OpenAI did in this case, but to get an idea of the kind of scaffolding and prompting strategy that one might want, have a look at this…

      jsnellView on HNParent context

  6. Claim: GPT-5-pro can prove new interesting mathematics 1 comment

    • Hypothesis: If you had ~1M dollar to burn, I think we should try setting up an AI agent to explore and try to invent new mathematics. It turns out agents can get an IMO gold with Gemini 2.5 Pro production model only. Therefore I suspect a swarm of agents burning through tokens…

      starchild3001View on HNParent context

  7. Let's properly analyze an AI article for once 1 comment

    • It is possible to take this too far, though - consider the OpenAI IMO proofs[1], for instance, and compare them to Gemini's.[2] [1] https://github.com/aw31/openai-imo-2025-proofs [2] https://arxiv.org/pdf/2507.15855 Appendix A

      986aignanView on HNParent context

  8. GPT-5 1 comment

    • IMO is not breakthrough, if you craft proper prompts you can excel imo with 2.5 Pro. Paper : https://arxiv.org/abs/2507.15855. Google just put whole computational power with very high quality data. It was test-time scaling. Why it didn't solve problem 6 as well? Yes, it was…

      demirbey05View on HNParent context

  9. Gemini with Deep Think achieves gold-medal standard at the IMO 1 comment

    • This paper [1] shows that gemini pro 2.5 without data-contamination and some minimal prompting (llm orchestration) can solve problems 1-5 on imo 2025. [1] https://arxiv.org/pdf/2507.15855 For problems with 2 page solutions, the search space of solutions is likely limited, and…

      bwfan123View on HNParent context

Submitted to Hacker News

Connected through shared HN threads, not inferred similarity or a recommendation.