Taiji Suzuki

Negative curvature can amplify PWGF approximation errors, but subsequent positive curvature can recover accuracy over growing time horizons, enabling stable optimization in non-convex landscapes.

On-policy exploration in RL for Transformers is provably superior to offline methods for hierarchical reasoning, achieving minimax-optimal rates where offline methods stall at logarithmic regret.

Generative transport reveals a hidden time-reflection symmetry in interpolants, enabling exact, non-parametric two-sample testing without density estimation.

Masked generative models can break the linear evaluation barrier during parallel decoding without ever needing to fully reconstruct the underlying token dependency graph.

ZF2ST achieves strong testing power for detecting distributional differences while maintaining rigorous statistical validity, outperforming traditional methods.

Takeuchi's Information Criterion (TIC) accurately predicts DNN generalization gaps, but only when models operate near the Neural Tangent Kernel (NTK) regime.