Yudong Chen

LLM pre-training power laws are not fixed but dynamically shaped by learning-rate and batch-size schedules, with a novel forcing-memory surrogate accurately predicting loss across these regimes.

Multi-action weakly-coupled MDPs can shatter the theoretical $O(1/\sqrt{N})$ performance barrier to reach an $O(1/N)$ optimality gap simply by ditching heuristic priority ordering in favor of locally linear mean-field dynamics.

Averaging outputs from multiple models can completely nullify watermark detection, raising critical questions about the future of AI text attribution.