3 citations · 6 across the 9 of their papers we have counts for
9 papers
PIPES: Securing Agent Perception with Provenance and Priors
Sanjay Kariyappa, Severin Klingler, G. Edward Suh
Tool-using agents consume external data from sources with different levels of trust, yet tool responses rarely identify who produced each component or what it should convey. We sho…
Where Instruction Hierarchy Breaks: Diagnosing and Repairing Failures in Reasoning Language Models
Sanjay Kariyappa, G. Edward Suh
Reasoning language models deployed in agentic workflows must follow an instruction hierarchy: when instructions from different sources conflict, the model should obey the highest-p…
Architecting Secure AI Agents: Perspectives on System-Level Defenses Against Indirect Prompt Injection Attacks
Chong Xiang, Drew Zagieboylo, Shaona Ghosh +5
AI agents, predominantly powered by large language models (LLMs), are vulnerable to indirect prompt injection, in which malicious instructions embedded in untrusted data can trigge…
ReasoningBomb: A Stealthy Denial-of-Service Attack by Inducing Pathologically Long Reasoning in Large Reasoning Models
Xiaogeng Liu, Xinyan Wang, Yechao Zhang +5
Large reasoning models (LRMs) extend large language models with explicit multi-step reasoning traces, but this capability introduces a new class of prompt-induced inference-time de…
Mitigating Indirect Prompt Injection via Instruction-Following Intent Analysis
Mintong Kang, Chong Xiang, Sanjay Kariyappa +3
Indirect prompt injection attacks (IPIAs), where large language models (LLMs) follow malicious instructions hidden in input data, pose a critical threat to LLM-powered agents. In t…
Stronger Enforcement of Instruction Hierarchy via Augmented Intermediate Representations
Sanjay Kariyappa, G. Edward Suh
Prompt injection attacks are a critical security vulnerability in large language models (LLMs), allowing attackers to hijack model behavior by injecting malicious instructions with…
Information Flow Control in Machine Learning through Modular Model Architecture
Trishita Tiwari, Suchin Gururangan, Chuan Guo +7
In today's machine learning (ML) models, any part of the training data can affect the model output. This lack of control for information flow from training data to model output is…
Bounding the Invertibility of Privacy-preserving Instance Encoding using Fisher Information
Kiwan Maeng, Chuan Guo, Sanjay Kariyappa +1
Privacy-preserving instance encoding aims to encode raw data as feature vectors without revealing their privacy-sensitive information. When designed properly, these encodings can b…
Measuring and Controlling Split Layer Privacy Leakage Using Fisher Information
Kiwan Maeng, Chuan Guo, Sanjay Kariyappa +1
Split learning and inference propose to run training/inference of a large model that is split across client devices and the cloud. However, such a model splitting imposes privacy c…