all AI news
A Geometric Notion of Causal Probing
Feb. 27, 2024, 5:50 a.m. | Cl\'ement Guerner, Anej Svete, Tianyu Liu, Alexander Warstadt, Ryan Cotterell
cs.CL updates on arXiv.org arxiv.org
Abstract: The linear subspace hypothesis (Bolukbasi et al., 2016) states that, in a language model's representation space, all information about a concept such as verbal number is encoded in a linear subspace. Prior work has relied on auxiliary classification tasks to identify and evaluate candidate subspaces that might give support for this hypothesis. We instead give a set of intrinsic criteria which characterize an ideal linear concept subspace and enable us to identify the subspace using …
abstract arxiv classification concept cs.cl hypothesis identify information language language model linear notion prior representation space support tasks type verbal work
More from arxiv.org / cs.CL updates on arXiv.org
Jobs in AI, ML, Big Data
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
Reporting & Data Analytics Lead (Sizewell C)
@ EDF | London, GB
Data Analyst
@ Notable | San Mateo, CA