← all papers · overview

Shattered Compositionality: Counterintuitive Learning Dynamics Of Transformers For Arithmetic

Abstract

Large language models (LLMs) often exhibit unexpected errors or unintended behavior, even at scale. While recent work reveals the discrepancy between LLMs and humans in skill compositions, the learning dynamics of skill compositions and the underlying cause of non-human behavior remain elusive. In this study, we investigate the mechanism of learning dynamics by training transformers on synthetic a

Related papers

Ranked by semantic similarity — how closely each paper's abstract matches this one (100% = near-identical topic).