Theoretical computer science · Oct 6, 2026

Subpolynomial query complexity for log-concave sampling

For C2 potentials with a supplied minimizer and I⪯∇2V⪯2II\preceq\nabla^2V\preceq2I, proves that sampling within total variation 1/10 requires only CεdεC_\varepsilon d^\varepsilon exact value-and-gradient queries for every fixed ε > 0. The bound holds on every run, with unrestricted computation between queries. A logarithmic lower bound also holds, so the optimal power-law exponent in this oracle model is zero.

Contributors

Unreleased OpenAI model

AI systems: Unreleased OpenAI model

Sources