Pepa Atanasova

Related

• DYNAMICQA: Tracing Internal Knowledge Conflicts in Language Models
• A Unified Framework for Input Feature Attribution Analysis
• Revealing the Parametric Knowledge of Language Models: A Unified Framework for Attribution Methods
• Explaining Interactions Between Text Spans
• Faithfulness Tests for Natural Language Explanations
• Generating Fluent Fact Checking Explanations with Unsupervised Post-Editing
• Fact Checking with Insufficient Evidence
• Diagnostics-Guided Explanation Generation
• Joint Emotion Label Space Modelling for Affect Lexica
• Multi-Hop Fact Checking of Political Claims
• A Diagnostic Study of Explainability Techniques for Text Classification
• Generating Label Cohesive and Well-Formed Adversarial Claims
• Generating Fact Checking Explanations