Work

Are Language Models Worse than Humans at Following Prompts? It’s Complicated

Year: 2023
Type: conference-paper
Abstract: Prompts have been the center of progress in advancing language models' zero-shot and fewshot performance.However, recent work finds that models can perform surprisingly well when given intentionally irrelevant or misleading prompts.Such results may be interpreted as evidence that model behavior is not "human like".In this study, we challenge a central assumption in such work: that humans would perform badly when given pathological instructions.We find that humans are able to reliably ignore irrelevant instructions and thus, like models, perform well on the underlying task despite an apparent lack of signal regarding the task they are being asked to do.However, when given deliberately misleading instructions, humans follow the instructions faithfully, whereas models do not.Our findings caution that future research should not idealize human behaviors as a monolith and should not train or evaluate models to mimic assumptions about these behaviors without first validating humans' behaviors empirically.
Institution: Brown University
Language: English
Open Access status: hybrid
OpenAlex ID: W4389519937
FWCI: 1.636
Cites: 17
Cited by: 28
Related to: 10