Learning the Wrong Lessons: Syntactic-Domain Spurious Correlations in Language Models
Author: Shaib, Chantal; Suriyakumar, Vinith M.; Sagun, Levent; Wallace, Byron C.; Ghassemi, Marzyeh Description: For an LLM to correctly respond to an instruction it must understand both the semantics and the domain (i.e., subject area) of a given task-instruction pair. However, syntax can also convey implicit information. Recent work shows that syntactic templates — frequent sequences of Part-of-Speech (PoS) tags — are prevalent in training data and often appear in model outputs. In this work we characterize syntactic templates, domain, and semantics in task-instruction pairs. We identify cases of spurious…
See more and a link to full text