PrivacySkills: How Privacy Guidance Shapes Source Selection in LLM Agents
cs.CR, cs.AI, cs.HC
Submitted: 2026-09-28
Updated: 2026-09-28
Terminology
Sources
- BiasBusters: Uncovering and Mitigating Tool Selection Bias in Large Language Models
- CI-Bench: Benchmarking Contextual Integrity of AI Assistants on Synthetic Data
- Securing AI Agents with Information-Flow Control
- Tool Preferences in Agentic LLMs are Unreliable
- What Affects the Stability of Tool Learning? An Empirical Study on the Robustness of Tool Learning Frameworks
- IDP-Bench: Benchmarking ability of LLMs to protect personal information in interdependent privacy contexts
- SoK: Agentic Skills -- Beyond Tool Use in LLM Agents
- ToolMinimize: Auditing and Rewriting LLM Agent Tool Calls to Minimize Privacy Exposure
- Can LLMs Keep a Secret? Testing Privacy Implications of Language Models via Contextual Integrity Theory
- gpt-oss-120b & gpt-oss-20b Model Card
- Agent Tools Orchestration Leaks More: Dataset, Benchmark, and Mitigation
- Progent: Securing AI Agents with Privilege Control
- PrivacyAlign: Contextual Privacy Alignment for LLM Agents
- Contextualized Privacy Defense for LLM Agents
- Agent Skills for Large Language Models: Architecture, Acquisition, Security, and the Path Forward
- When Lower Privileges Suffice: Investigating Over-Privileged Tool Selection in LLM Agents
- Nemotron-Cascade 2: Post-Training LLMs with Cascade RL and Multi-Domain On-Policy Distillation
- PrivacyPeek: Auditing What LLM-Based Agents Acquire, Not Just What They Say
- AgentDAM: Privacy Leakage Evaluation for Autonomous Web Agents
Related papers
- SoK: AI-Augmented Binary Reversing
- Relaxed Sender Anonymity for CBDC Interbank Settlement: A Zero-Knowledge Approach on Permissioned EVM
- Calibration-Family Overfit: Why Trusted Sabotage Monitors Don't Transfer Across Lineages
- Efficient Fuzzy PSI under One-Sided Assumptions
- Sealing the Audit-Runtime Gap for LLM Skills
- Token Composition: A Graph Based on EVM Logs