Sensitive-Topic Leakage Through LLM Routing Metadata: Measurement and Mitigation
cs.CR, cs.CL
Submitted: 2026-10-07
Updated: 2026-10-07
Code: https://github.com/luka-krixvon/routing-metadata-leakage
Terminology
Sources
- FrugalGPT: How to Use Large Language Models While Reducing Cost and Improving Performance
- Hybrid LLM: Cost-Efficient and Quality-Aware Query Routing
- AutoMix: Automatically Mixing Language Models
- RouteLLM: Learning to Route LLMs with Preference Data
- RouterBench: A Benchmark for Multi-LLM Routing System
- Register Bias in Complexity-Based Large Language Model Routing
- Privacy-Preserving LLMs Routing
- Whisper Leak: a side-channel attack on Large Language Models
- A Large-Scale Benchmark and Risk Assessment of Traffic Analysis Attacks on Cloud LLM Services
- From Privacy to Workflow Integrity: Communication-Graph Metadata in Autonomous Agent Interoperability
- Selection-Invariant Communication Compilers for Privacy-Aware Multi-Agent LLM Workflows
- AVEC: Bootstrapping Privacy for Local LLMs
- Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preference
- Task- and Session-Level Model Routing: A Common-Interface Hybrid Evaluation of Four Open-Source Routers Across Four Benchmarks
- Remote Timing Attacks on Efficient Language Model Inference
- RouteScan: A Non-Intrusive Approach to Auditing MoE LLMs Safety via Expert Routing Telemetry
- Hidden Risks of Jev: An Empirical Study of Security, Privacy, and Dual Use
- Equality of Opportunity in Supervised Learning
- Fair and Optimal Classification via Post-Processing
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