Picture for Shauli Ravfogel

Shauli Ravfogel

Counterfactual Generation from Language Models

Add code
Nov 11, 2024
Viaarxiv icon

GRADE: Quantifying Sample Diversity in Text-to-Image Models

Add code
Oct 29, 2024
Figure 1 for GRADE: Quantifying Sample Diversity in Text-to-Image Models
Figure 2 for GRADE: Quantifying Sample Diversity in Text-to-Image Models
Figure 3 for GRADE: Quantifying Sample Diversity in Text-to-Image Models
Figure 4 for GRADE: Quantifying Sample Diversity in Text-to-Image Models
Viaarxiv icon

Intrinsic Evaluation of Unlearning Using Parametric Knowledge Traces

Add code
Jun 17, 2024
Figure 1 for Intrinsic Evaluation of Unlearning Using Parametric Knowledge Traces
Figure 2 for Intrinsic Evaluation of Unlearning Using Parametric Knowledge Traces
Figure 3 for Intrinsic Evaluation of Unlearning Using Parametric Knowledge Traces
Figure 4 for Intrinsic Evaluation of Unlearning Using Parametric Knowledge Traces
Viaarxiv icon

On Affine Homotopy between Language Encoders

Add code
Jun 04, 2024
Figure 1 for On Affine Homotopy between Language Encoders
Figure 2 for On Affine Homotopy between Language Encoders
Figure 3 for On Affine Homotopy between Language Encoders
Figure 4 for On Affine Homotopy between Language Encoders
Viaarxiv icon

Language Imbalance Can Boost Cross-lingual Generalisation

Add code
Apr 11, 2024
Viaarxiv icon

What Changed? Converting Representational Interventions to Natural Language

Add code
Feb 17, 2024
Viaarxiv icon

MiMiC: Minimally Modified Counterfactuals in the Representation Space

Add code
Feb 16, 2024
Viaarxiv icon

Guiding LLM to Fool Itself: Automatically Manipulating Machine Reading Comprehension Shortcut Triggers

Add code
Oct 24, 2023
Viaarxiv icon

The Curious Case of Hallucinatory Unanswerablity: Finding Truths in the Hidden States of Over-Confident Large Language Models

Add code
Oct 18, 2023
Figure 1 for The Curious Case of Hallucinatory Unanswerablity: Finding Truths in the Hidden States of Over-Confident Large Language Models
Figure 2 for The Curious Case of Hallucinatory Unanswerablity: Finding Truths in the Hidden States of Over-Confident Large Language Models
Figure 3 for The Curious Case of Hallucinatory Unanswerablity: Finding Truths in the Hidden States of Over-Confident Large Language Models
Figure 4 for The Curious Case of Hallucinatory Unanswerablity: Finding Truths in the Hidden States of Over-Confident Large Language Models
Viaarxiv icon

LEACE: Perfect linear concept erasure in closed form

Add code
Jun 23, 2023
Viaarxiv icon