What Happens to a Dataset Transformed by a Projection-based Concept Removal Method?

arXiv:2403.16142v1 Announce Type: new
Abstract: We investigate the behavior of methods that use linear projections to remove information about a concept from a language representation, and we consider the question of what happens to a dataset transformed by such a method. A theoretical analysis and experiments on real-world and synthetic data show that these methods inject strong statistical dependencies into the transformed datasets. After applying such a method, the representation space is highly structured: in the transformed space, an instance …

abstract analysis arxiv behavior concept cs.ai cs.cl dataset information language linear projection question representation type world