Picture for Fan Yin

Fan Yin

Characterizing Truthfulness in Large Language Model Generations with Local Intrinsic Dimension

Add code
Feb 28, 2024
Viaarxiv icon

Contrastive Instruction Tuning

Add code
Feb 17, 2024
Viaarxiv icon

Prompt-Driven LLM Safeguarding via Directed Representation Optimization

Add code
Jan 31, 2024
Viaarxiv icon

Active Instruction Tuning: Improving Cross-Task Generalization by Training on Prompt Sensitive Tasks

Add code
Nov 01, 2023
Viaarxiv icon

Should I Stop or Should I Go: Early Stopping with Heterogeneous Populations

Add code
Jun 26, 2023
Figure 1 for Should I Stop or Should I Go: Early Stopping with Heterogeneous Populations
Figure 2 for Should I Stop or Should I Go: Early Stopping with Heterogeneous Populations
Figure 3 for Should I Stop or Should I Go: Early Stopping with Heterogeneous Populations
Figure 4 for Should I Stop or Should I Go: Early Stopping with Heterogeneous Populations
Viaarxiv icon

Did You Read the Instructions? Rethinking the Effectiveness of Task Definitions in Instruction Learning

Add code
Jun 01, 2023
Figure 1 for Did You Read the Instructions? Rethinking the Effectiveness of Task Definitions in Instruction Learning
Figure 2 for Did You Read the Instructions? Rethinking the Effectiveness of Task Definitions in Instruction Learning
Figure 3 for Did You Read the Instructions? Rethinking the Effectiveness of Task Definitions in Instruction Learning
Figure 4 for Did You Read the Instructions? Rethinking the Effectiveness of Task Definitions in Instruction Learning
Viaarxiv icon

Efficient Shapley Values Estimation by Amortization for Text Classification

Add code
May 31, 2023
Figure 1 for Efficient Shapley Values Estimation by Amortization for Text Classification
Figure 2 for Efficient Shapley Values Estimation by Amortization for Text Classification
Figure 3 for Efficient Shapley Values Estimation by Amortization for Text Classification
Figure 4 for Efficient Shapley Values Estimation by Amortization for Text Classification
Viaarxiv icon

Red Teaming Language Model Detectors with Language Models

Add code
May 31, 2023
Figure 1 for Red Teaming Language Model Detectors with Language Models
Figure 2 for Red Teaming Language Model Detectors with Language Models
Figure 3 for Red Teaming Language Model Detectors with Language Models
Figure 4 for Red Teaming Language Model Detectors with Language Models
Viaarxiv icon

Dynosaur: A Dynamic Growth Paradigm for Instruction-Tuning Data Curation

Add code
May 23, 2023
Figure 1 for Dynosaur: A Dynamic Growth Paradigm for Instruction-Tuning Data Curation
Figure 2 for Dynosaur: A Dynamic Growth Paradigm for Instruction-Tuning Data Curation
Figure 3 for Dynosaur: A Dynamic Growth Paradigm for Instruction-Tuning Data Curation
Figure 4 for Dynosaur: A Dynamic Growth Paradigm for Instruction-Tuning Data Curation
Viaarxiv icon

CleanCLIP: Mitigating Data Poisoning Attacks in Multimodal Contrastive Learning

Add code
Mar 08, 2023
Figure 1 for CleanCLIP: Mitigating Data Poisoning Attacks in Multimodal Contrastive Learning
Figure 2 for CleanCLIP: Mitigating Data Poisoning Attacks in Multimodal Contrastive Learning
Figure 3 for CleanCLIP: Mitigating Data Poisoning Attacks in Multimodal Contrastive Learning
Figure 4 for CleanCLIP: Mitigating Data Poisoning Attacks in Multimodal Contrastive Learning
Viaarxiv icon