Decentralized SGD under Heavy-Tailed Noise: Optimal Convergence Rates and the Role of Gradient Clipping

Researchers propose G²MLP, a graph geometry-aware MLP distillation method that addresses spectral failure modes in GNN-to-MLP knowledge transfer using Ollivier-Ricci curvature.

Research paper identifies systematic language sensitivity in vision-language-action models where minor phrasing changes cause performance swings up to 61 points, and proposes rule-based rephrasing method that improves success rates by 16-27%.

Researchers propose Exploration-Distillation (ExpDis), a method that decouples exploration from optimization in RLVR training, improving mathematical reasoning performance across seven benchmarks.

EngramEdit enables precise LLM knowledge updates via conditional memory, achieving 3x baseline accuracy in chain-of-thought reasoning while preserving unrelated knowledge.

Mathematicians are struggling to understand 372 AI-generated proofs from OpenAI, including a proof of the long-unsolved Unique Games Conjecture, which appear superhuman but incomprehensible to humans.