all AI news
Know When To Stop: A Study of Semantic Drift in Text Generation
April 9, 2024, 4:50 a.m. | Ava Spataru, Eric Hambro, Elena Voita, Nicola Cancedda
cs.CL updates on arXiv.org arxiv.org
Abstract: In this work, we explicitly show that modern LLMs tend to generate correct facts first, then "drift away" and generate incorrect facts later: this was occasionally observed but never properly measured. We develop a semantic drift score that measures the degree of separation between correct and incorrect facts in generated texts and confirm our hypothesis when generating Wikipedia-style biographies. This correct-then-incorrect generation pattern suggests that factual accuracy can be improved by knowing when to stop …
abstract arxiv cs.cl drift facts generate llms modern semantic show study text text generation type work
More from arxiv.org / cs.CL updates on arXiv.org
Jobs in AI, ML, Big Data
AI Research Scientist
@ Vara | Berlin, Germany and Remote
Data Architect
@ University of Texas at Austin | Austin, TX
Data ETL Engineer
@ University of Texas at Austin | Austin, TX
Lead GNSS Data Scientist
@ Lurra Systems | Melbourne
Senior Machine Learning Engineer (MLOps)
@ Promaton | Remote, Europe
AI Engineering Manager
@ M47 Labs | Barcelona, Catalunya [Cataluña], Spain