Evaluating explainability in language classification models: A unified framework incorporating feature attribution methods and key factors affecting faithfulness

Tahereh Dehdarirad

doi:10.1016/j.dim.2025.100101

ScienceGate Book Chapters

JOURNAL ARTICLE

Evaluating explainability in language classification models: A unified framework incorporating feature attribution methods and key factors affecting faithfulness

Tahereh Dehdarirad

Year: 2025 Journal: Data and Information Management Vol: 9 (4)Pages: 100101-100101 Publisher: De Gruyter

DOI: 10.1016/j.dim.2025.100101

Get Full-Text PDF Get Analytical Report

Keywords:

Key (lock) Computer science Feature (linguistics) Attribution Artificial intelligence Natural language processing Linguistics Psychology Social psychology Computer security

Metrics

Cited By

24.10

FWCI (Field Weighted Citation Impact)

Refs

0.99

Citation Normalized Percentile

Is in top 1%

Is in top 10%

Citation History

Topics

Explainable Artificial Intelligence (XAI)

Physical Sciences → Computer Science → Artificial Intelligence

Topic Modeling

Physical Sciences → Computer Science → Artificial Intelligence

Adversarial Robustness in Machine Learning

Physical Sciences → Computer Science → Artificial Intelligence

Evaluating explainability in language classification models: A unified framework incorporating feature attribution methods and key factors affecting faithfulness

Metrics

Citation History

Topics

Related Documents

Counterfactuals As a Means for Evaluating Faithfulness of Attribution Methods in Autoregressive Language Models

Empirical Analysis of Methods for Evaluating Faithfulness of Explanations by Feature Attribution

TextFocus: Assessing the Faithfulness of Feature Attribution Methods Explanations in Natural Language Processing

Revealing the Parametric Knowledge of Language Models: A Unified Framework for Attribution Methods

IntActEval: A Continuous Framework for Quantitatively Evaluating Faithfulness of Feature Visualization Methods