Loose Lips: Understanding Privacy Leakage in Agentic Dialogue
cs.CL, cs.AI
Submitted: 2026-10-07
Updated: 2026-10-07
Code: https://github.com/soummyaah/LooseLips
Sources
- PrivAct: Internalizing Contextual Privacy Preservation via Multi-Agent Preference Training
- Do Thinking Tokens Help or Trap? Towards More Efficient Large Reasoning Model
- Operationalizing Contextual Integrity in Privacy-Conscious Assistants
- MAGPIE: A dataset for Multi-AGent contextual PrIvacy Evaluation
- TAMAS: Benchmarking Adversarial Risks in Multi-Agent LLM Systems
- Can LLMs Keep a Secret? Testing Privacy Implications of Language Models via Contextual Integrity Theory
- CIMemories: A Compositional Benchmark for Contextual Integrity of Persistent Memory in LLMs
- PrivacyBench: A Conversational Benchmark for Evaluating Privacy in Personalized AI
- LeakAgent: RL-based Red-teaming Agent for LLM Privacy Leakage
- PAC-BENCH: Evaluating Multi-Agent Collaboration under Privacy Constraints
- Qwen3 Technical Report
- AgentLeak: A Benchmark for Internal-Channel Privacy Leakage in Multi-Agent LLM Systems
- CalBench: Evaluating Coordination-Privacy Trade-offs in Multi-Agent LLMs
Related papers
- Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
- Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
- Subliminal Steering: Stronger Encoding of Hidden Signals
- MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
- The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
- Untangling the Mechanisms of Misleading Context in Medical Question Answering