Picture for Sahar Abdelnabi

Sahar Abdelnabi

Hypothesizing Missing Causal Variables with LLMs

Add code
Sep 04, 2024
Viaarxiv icon

Dataset and Lessons Learned from the 2024 SaTML LLM Capture-the-Flag Competition

Add code
Jun 12, 2024
Viaarxiv icon

Are you still on track!? Catching LLM Task Drift with Activations

Add code
Jun 02, 2024
Viaarxiv icon

Can LLMs Separate Instructions From Data? And What Do We Even Mean By That?

Add code
Mar 11, 2024
Viaarxiv icon

Exploring Value Biases: How LLMs Deviate Towards the Ideal

Add code
Feb 21, 2024
Viaarxiv icon

LLM-Deliberation: Evaluating LLMs with Interactive Multi-Agent Negotiation Games

Add code
Sep 29, 2023
Viaarxiv icon

More than you've asked for: A Comprehensive Analysis of Novel Prompt Injection Threats to Application-Integrated Large Language Models

Add code
Feb 23, 2023
Viaarxiv icon

Fact-Saboteurs: A Taxonomy of Evidence Manipulation Attacks against Fact-Verification Systems

Add code
Sep 07, 2022
Figure 1 for Fact-Saboteurs: A Taxonomy of Evidence Manipulation Attacks against Fact-Verification Systems
Figure 2 for Fact-Saboteurs: A Taxonomy of Evidence Manipulation Attacks against Fact-Verification Systems
Figure 3 for Fact-Saboteurs: A Taxonomy of Evidence Manipulation Attacks against Fact-Verification Systems
Figure 4 for Fact-Saboteurs: A Taxonomy of Evidence Manipulation Attacks against Fact-Verification Systems
Viaarxiv icon

Open-Domain, Content-based, Multi-modal Fact-checking of Out-of-Context Images via Online Resources

Add code
Dec 07, 2021
Figure 1 for Open-Domain, Content-based, Multi-modal Fact-checking of Out-of-Context Images via Online Resources
Figure 2 for Open-Domain, Content-based, Multi-modal Fact-checking of Out-of-Context Images via Online Resources
Figure 3 for Open-Domain, Content-based, Multi-modal Fact-checking of Out-of-Context Images via Online Resources
Figure 4 for Open-Domain, Content-based, Multi-modal Fact-checking of Out-of-Context Images via Online Resources
Viaarxiv icon

"What's in the box?!": Deflecting Adversarial Attacks by Randomly Deploying Adversarially-Disjoint Models

Add code
Mar 09, 2021
Figure 1 for "What's in the box?!": Deflecting Adversarial Attacks by Randomly Deploying Adversarially-Disjoint Models
Figure 2 for "What's in the box?!": Deflecting Adversarial Attacks by Randomly Deploying Adversarially-Disjoint Models
Figure 3 for "What's in the box?!": Deflecting Adversarial Attacks by Randomly Deploying Adversarially-Disjoint Models
Figure 4 for "What's in the box?!": Deflecting Adversarial Attacks by Randomly Deploying Adversarially-Disjoint Models
Viaarxiv icon