2 papers
cs.AI2026
Unveiling the Reasoning Process of Large Language Models
Junjie Zhang, Zhen Shen, Xisong Dong +1
Large language models often reason beyond surface tokens, but the internal stage at which token-level information becomes abstract relational structure remains unclear. We investig…
cs.AI2026
Grokking From Abstraction to Intelligence
Junjie Zhang, Zhen Shen, Gang Xiong +1
Grokking in modular arithmetic has established itself as the quintessential fruit fly experiment, serving as a critical domain for investigating the mechanistic origins of model ge…