The Hidden Costs of Domain Fine-Tuning: Pii-Bearing Data Degrades Safety and Increases Leakage
arXiv 2603.00061•d71508205dd81d88559688eca10eb3ab5e51f8a7c575b6318a37e5615ed68813
LLM safetyPII leakageRAG poisoningReverse CAPTCHATEE (SGX)TalariaThreat modelingThreatFormer-IDSZK-PoPadversarial MLagentic AI skillsbehavioral biometrics (keystroke)confidential inferencedomain fine-tuningformal analysisintrusion detectioninvisible Unicodemetadata poisoningmultimodal securityprompt injectionskill marketplace securitysupply chain attackstoken-inference attackszero-day generalizationzero-knowledge proofs
Paper metadata
- arXiv ID
- 2603.00061
- Version
- Not specified by this published record
- Category
- Computer Science — Cryptography and Security (cs.CR)
The PDF link points to arxiv.org. Baitaphish does not expose a private stored PDF.
Evidence and limitations
- Source ID
- arxiv_cs_cr
- Record identifier
- d71508205dd81d88559688eca10eb3ab5e51f8a7c575b6318a37e5615ed68813
- Enrichment time
- 2026-03-04T19:46:30Z
- AI-assisted enrichment
- Yes
This record may overlap with other records. Its enrichment can be incomplete or wrong, and machine assistance was used. Validate consequential decisions against the linked source and your own environment.