3 papers
cs.LG2025
Honesty over Accuracy: Trustworthy Language Models through Reinforced Hesitation
Mohamad Amin Mohamadi, Tianhao Wang, Zhiyuan Li
Modern language models fail a fundamental requirement of trustworthy intelligence: knowing when not to answer. Despite achieving impressive accuracy on benchmarks, these models pro…
cs.LG2025
Adam Exploits -geometry of Loss Landscape via Coordinate-wise Adaptivity
Shuo Xie, Mohamad Amin Mohamadi, Zhiyuan Li
Adam outperforms SGD when training language models. Yet this advantage is not well-understood theoretically -- previous convergence analysis for Adam and SGD mainly focuses on the…
cs.LG2024
Why Do You Grok? A Theoretical Analysis of Grokking Modular Addition
Mohamad Amin Mohamadi, Zhiyuan Li, Lei Wu +1
We present a theoretical explanation of the ``grokking'' phenomenon, where a model generalizes long after overfitting,for the originally-studied problem of modular addition. First,…