Picture for Oana-Maria Camburu

Oana-Maria Camburu

The Probabilities Also Matter: A More Faithful Metric for Faithfulness of Free-Text Explanations in Large Language Models

Apr 04, 2024
Viaarxiv icon

Identifying Linear Relational Concepts in Large Language Models

Add code
Nov 15, 2023
Viaarxiv icon

Using Natural Language Explanations to Improve Robustness of In-context Learning for Natural Language Inference

Add code
Nov 13, 2023
Viaarxiv icon

KNOW How to Make Up Your Mind! Adversarially Detecting and Alleviating Inconsistencies in Natural Language Explanations

Add code
Jun 05, 2023
Figure 1 for KNOW How to Make Up Your Mind! Adversarially Detecting and Alleviating Inconsistencies in Natural Language Explanations
Figure 2 for KNOW How to Make Up Your Mind! Adversarially Detecting and Alleviating Inconsistencies in Natural Language Explanations
Figure 3 for KNOW How to Make Up Your Mind! Adversarially Detecting and Alleviating Inconsistencies in Natural Language Explanations
Figure 4 for KNOW How to Make Up Your Mind! Adversarially Detecting and Alleviating Inconsistencies in Natural Language Explanations
Viaarxiv icon

Faithfulness Tests for Natural Language Explanations

Add code
May 29, 2023
Figure 1 for Faithfulness Tests for Natural Language Explanations
Figure 2 for Faithfulness Tests for Natural Language Explanations
Figure 3 for Faithfulness Tests for Natural Language Explanations
Figure 4 for Faithfulness Tests for Natural Language Explanations
Viaarxiv icon

SPARSEFIT: Few-shot Prompting with Sparse Fine-tuning for Jointly Generating Predictions and Natural Language Explanations

May 23, 2023
Figure 1 for SPARSEFIT: Few-shot Prompting with Sparse Fine-tuning for Jointly Generating Predictions and Natural Language Explanations
Figure 2 for SPARSEFIT: Few-shot Prompting with Sparse Fine-tuning for Jointly Generating Predictions and Natural Language Explanations
Figure 3 for SPARSEFIT: Few-shot Prompting with Sparse Fine-tuning for Jointly Generating Predictions and Natural Language Explanations
Figure 4 for SPARSEFIT: Few-shot Prompting with Sparse Fine-tuning for Jointly Generating Predictions and Natural Language Explanations
Viaarxiv icon

Logical Reasoning for Natural Language Inference Using Generated Facts as Atoms

May 22, 2023
Figure 1 for Logical Reasoning for Natural Language Inference Using Generated Facts as Atoms
Figure 2 for Logical Reasoning for Natural Language Inference Using Generated Facts as Atoms
Figure 3 for Logical Reasoning for Natural Language Inference Using Generated Facts as Atoms
Figure 4 for Logical Reasoning for Natural Language Inference Using Generated Facts as Atoms
Viaarxiv icon

Counter-GAP: Counterfactual Bias Evaluation through Gendered Ambiguous Pronouns

Add code
Feb 11, 2023
Figure 1 for Counter-GAP: Counterfactual Bias Evaluation through Gendered Ambiguous Pronouns
Figure 2 for Counter-GAP: Counterfactual Bias Evaluation through Gendered Ambiguous Pronouns
Figure 3 for Counter-GAP: Counterfactual Bias Evaluation through Gendered Ambiguous Pronouns
Figure 4 for Counter-GAP: Counterfactual Bias Evaluation through Gendered Ambiguous Pronouns
Viaarxiv icon

Rationalizing Predictions by Adversarial Information Calibration

Add code
Jan 15, 2023
Figure 1 for Rationalizing Predictions by Adversarial Information Calibration
Figure 2 for Rationalizing Predictions by Adversarial Information Calibration
Figure 3 for Rationalizing Predictions by Adversarial Information Calibration
Figure 4 for Rationalizing Predictions by Adversarial Information Calibration
Viaarxiv icon

Explaining Chest X-ray Pathologies in Natural Language

Add code
Jul 09, 2022
Figure 1 for Explaining Chest X-ray Pathologies in Natural Language
Figure 2 for Explaining Chest X-ray Pathologies in Natural Language
Figure 3 for Explaining Chest X-ray Pathologies in Natural Language
Figure 4 for Explaining Chest X-ray Pathologies in Natural Language
Viaarxiv icon