4 papers
Environment Steering: Using Data Flow Control to Improve Agent Utility and Safety
Charlie Summers, Prajwal Raghunath, Aaditya Pai +4
LLM agents can make unsafe tool calls even when instructed to behave safely. Existing defenses constrain agents before execution, modify tool inputs/outputs, or rely on LLM judges;…
PARSE: Provenance-Aware Retrieval Sanitization for Professional Domain LLM Agents
Aaditya Pai
Prompt injection defenses evaluated on synthetic benchmarks do not generalize to real enterprise documents, which are longer, denser, and interleave legitimate authority language w…
Evaluating Prompting-Based Defenses Against Domain-Camouflaged Injection Attacks
Aaditya Pai
Domain-camouflaged injection attacks embed malicious instructions in retrieved content using domain-appropriate vocabulary, evading standard detectors that rely on syntactic inject…
Blind Spots in the Guard: How Domain-Camouflaged Injection Attacks Evade Detection in Multi-Agent LLM Systems
Aaditya Pai
Injection detectors deployed to protect LLM agents are calibrated on static, template-based payloads that announce themselves as override directives. We identify a systematic blind…