Work

Evaluating Concept Discovery Methods for Sensitive Attributes in Language Models

Year: 2025

Type: conference-paper

Abstract: This paper examines how to improve interpretability of language models in the context of fairness.While traditional concept learning focuses on identifying the most important concepts for a task, this study explores how to locate the representation of sensitive attributes in pretrained language models.We address challenges such as the potential low importance and sparsity of sensitive attributes in training data, and the limited amount of labeled data for this purpose.Our experiments evaluate potential methods to obtain such identity concepts, considering factors like label sparsity, generalizability, and the influence of different language models on the representation of sensitive attributes.

Source: European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN)

DOI: 10.14428/esann/2025.es2025-150

Authors: Sarah Schröder, Alexander Schulz, Barbara Hammer

Institution: Bielefeld University

Language: English

Topic: Explainable Artificial Intelligence (XAI)

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

SDG: Reduced inequalities

Open Access status: bronze

OpenAlex ID: W4409480950

FWCI:

Cites:

Cited by:

Related to: 10