collaborators

8 papers

cs.CR2026

Undetectable Backdoors in Model Parameters: Hiding Sparse Secrets in High Dimensions

Sarthak Choudhary, Atharv Singh Patlan, Nils Palumbo +3

We present Sparse Backdoor, a supply-chain attack that plants a provably undetectable backdoor in pre-trained image classifiers, including convolutional networks and Vision Transfo…

cs.CR2026

Through the Stealth Lens: Attention-Aware Defenses Against Poisoning in RAG

Sarthak Choudhary, Nils Palumbo, Ashish Hooda +2

Retrieval-augmented generation (RAG) systems are vulnerable to attacks that inject poisoned passages into the retrieved context, even at low corruption rates. We show that existing…

cs.CR2026

Agent Security is a Systems Problem

Mihai Christodorescu, Earlence Fernandes, Ashish Hooda +11

We take the position that agent security must be approached as a systems problem: the AI model powering the agent must be treated as an untrusted component, and security invariants…

cs.CR2026

Formal Policy Enforcement for Real-World Agentic Systems

Nils Palumbo, Sarthak Choudhary, Jihye Choi +3

Security policy enforcement in contemporary agentic systems predominantly consists of embedding natural-language policies within an agent's system prompt and delegating compliance…

cs.CR2026

Dependency-Aware Privacy for Multi-turn Agents

Divyam Anshumaan, Sarthak Choudhary, Nils Palumbo +1

LLM agents release private data across multi-service interactions. Existing prompt sanitizers based on metric differential privacy treat each release independently, so adversaries…

cs.CR2025

How Not to Detect Prompt Injections with an LLM

Sarthak Choudhary, Divyam Anshumaan, Nils Palumbo +1

LLM-integrated applications and agents are vulnerable to prompt injection attacks, where adversaries embed malicious instructions within seemingly benign input data to manipulate t…