Search papers, labs, and topics across Lattice.
Riemann-Bench is introduced as a private benchmark of 25 expert-curated, research-level mathematics problems designed to evaluate AI systems beyond the scope of olympiad-style competitions. The problems, authored and verified by mathematicians, require weeks to solve and have unique, closed-form solutions. Evaluation of frontier models as unconstrained research agents reveals performance below 10%, highlighting a significant gap in research-level mathematical reasoning.
Today's best AI models are humbled by research-level math problems, scoring below 10% despite excelling at olympiad-style competitions.
Recent AI systems have achieved gold-medal-level performance on the International Mathematical Olympiad, demonstrating remarkable proficiency at competition-style problem solving. However, competition mathematics represents only a narrow slice of mathematical reasoning: problems are drawn from limited domains, require minimal advanced machinery, and can often reward insightful tricks over deep theoretical knowledge. We introduce \bench{}, a private benchmark of 25 expert-curated problems designed to evaluate AI systems on research-level mathematics that goes far beyond the olympiad frontier. Problems are authored by Ivy League mathematics professors, graduate students, and PhD-holding IMO medalists, and routinely took their authors weeks to solve independently. Each problem undergoes double-blind verification by two independent domain experts who must solve the problem from scratch, and yields a unique, closed-form solution assessed by programmatic verifiers. We evaluate frontier models as unconstrained research agents, with full access to coding tools, search, and open-ended reasoning, using an unbiased statistical estimator computed over 100 independent runs per problem. Our results reveal that all frontier models currently score below 10\%, exposing a substantial gap between olympiad-level problem solving and genuine research-level mathematical reasoning. By keeping the benchmark fully private, we ensure that measured performance reflects authentic mathematical capability rather than memorization of training data.
