Picture for Zhenlan Ji

Zhenlan Ji

SelfDefend: LLMs Can Defend Themselves against Jailbreaking in a Practical Manner

Add code
Jun 08, 2024
Viaarxiv icon

Testing and Understanding Erroneous Planning in LLM Agents through Synthesized User Inputs

Add code
Apr 27, 2024
Viaarxiv icon

InstructTA: Instruction-Tuned Targeted Attack for Large Vision-Language Models

Add code
Dec 04, 2023
Figure 1 for InstructTA: Instruction-Tuned Targeted Attack for Large Vision-Language Models
Figure 2 for InstructTA: Instruction-Tuned Targeted Attack for Large Vision-Language Models
Figure 3 for InstructTA: Instruction-Tuned Targeted Attack for Large Vision-Language Models
Figure 4 for InstructTA: Instruction-Tuned Targeted Attack for Large Vision-Language Models
Viaarxiv icon

Benchmarking and Explaining Large Language Model-based Code Generation: A Causality-Centric Approach

Add code
Oct 10, 2023
Viaarxiv icon

Enabling Runtime Verification of Causal Discovery Algorithms with Automated Conditional Independence Reasoning (Extended Version)

Add code
Sep 11, 2023
Viaarxiv icon

Causality-Aided Trade-off Analysis for Machine Learning Fairness

Add code
May 22, 2023
Viaarxiv icon