emergent capabilities 1interpretability 1nucleation theory 1training dynamics 1transformer circuit analysis 1
From the 1 of 2 linked papers with an AI index.
2 papers
cs.LG2026
The Kinetics of Training: A Driven-Nucleation Rate Law for Emergence, Plasticity Loss, and Circuit Control in Language Models
Lei Dong
The paper proposes a driven‑nucleation rate law describing how language model capabilities emerge only when an entire circuit aligns in a single stochastic event, and shows that re…
cs.CL2026
Better Starts, Better Ends: Bootstrapped Iterative Self-Reasoning Distillation for Compressed Reasoning
Leichao Dong, Dongxu Zhang, Yiding Sun +4
Large reasoning models often solve problems through long chain-of-thought (CoT) traces, yet much of this computation is spent on redundant derivations, repeated self-verification,…