AI Model Achieves Gold-Medal Performance on International Mathematics Olympiad Problems
Researchers from MiniMax have introduced MaxProof, a test-time scaling framework for mathematical proof that scores 35/42 on IMO 2025 and 36/42 on USAMO 2026, exceeding the human gold-medal threshold on both competitions. The system builds on a model called MiniMax-M3, which is trained in three proof-oriented capabilities — generation, verification, and critique-conditioned repair — and uses tournament selection at inference time to select a final proof from a population of candidates. The results represent a significant milestone in AI mathematical reasoning, demonstrating that competition-level formal proof is now within reach of automated systems.
MaxProof is a population-level test-time scaling framework developed by MiniMax as part of their M3 model series, designed to tackle competition-level mathematical proof. The underlying MiniMax-M3 model is trained with three distinct capabilities: proof generation, proof verification, and critique-conditioned proof repair, unified into a single released model using a defense-in-depth generative verifier engineered for a low false-positive rate. At inference time, MaxProof deploys the model simultaneously as generator, verifier, refiner, and ranker, searching over a population of candidate proofs before selecting a final answer via tournament selection. This approach yielded scores of 35/42 on IMO 2025 and 36/42 on USAMO 2026, both surpassing the human gold-medal threshold for those competitions. The work was submitted to arXiv on June 11, 2026, and has not yet undergone formal peer review. The framework represents a notable advance in applying reinforcement learning and generative verification to high-difficulty mathematical reasoning tasks.
What's missing
The paper has not yet undergone peer review, so independent validation of the benchmark scores is absent. Key open questions include: whether the IMO 2025 and USAMO 2026 problems were fully withheld from training data (data contamination risk), the exact definition of the 'human gold-medal threshold' used for comparison, computational cost and inference time required to run MaxProof at scale, and whether the proofs produced are formally verified in a proof assistant or assessed by human judges.
What different sources said
- arXiv cs.AICenter
MaxProof: Scaling Mathematical Proof with Generative-Verifier RL and Population-Level Test-Time Scaling
Related
Gut Bacteria Enzyme Found to Break Down Heat-Processed Food Compounds, Producing Novel Biogenic Amines
Researchers have discovered that an enzyme in common gut bacteria can degrade N-epsilon-carboxymethyllysine (CML), a compound formed during thermal food processing, producing previously unknown biogenic amines. The enzyme, ornithine decarboxylase SpeC from enterobacteria, acts on CML and related modified lysine derivatives through a low-level 'underground' catalytic activity. This finding suggests a previously unrecognized communication axis between thermally processed dietary compounds and gut microbial physiology, with potential implications for host health.
Full-Length Gene Sequencing Reveals Two Distinct Bacterial Communities in Black-Legged Ticks Expanding Into Canada
Researchers used Oxford Nanopore full-length 16S rRNA gene sequencing to characterize the microbiome of Ixodes scapularis black-legged ticks collected in Nova Scotia, Canada, distinguishing between tick-adapted bacteria and environmentally acquired bacteria. The study comes as I. scapularis — the primary vector of Lyme disease — is rapidly expanding northward into Canada due to climate change. The findings suggest that environmentally derived bacteria in tick microbiomes are not mere contamination, which has implications for how tick microbiome data is collected and interpreted across surveillance studies.
Study Identifies Metabolic Link Between Cell Envelope Stress and Biofilm Formation in Bacteria
Researchers have discovered that the metabolite acetyl-CoA directly inhibits enzymes that degrade the bacterial signaling molecule c-di-GMP, connecting cell envelope biosynthesis stress to biofilm formation in Pseudomonas aeruginosa. The study found that sub-inhibitory concentrations of antibiotics targeting early peptidoglycan biosynthesis — but not other antibiotic classes — elevate c-di-GMP levels by reducing phosphodiesterase activity, with acetyl-CoA competing for the enzyme active site. Because the relevant enzyme domain is broadly conserved across bacterial species, this checkpoint mechanism may be widespread and could have implications for understanding antibiotic-induced biofilm responses.