Label Leakage Attacks in Machine Unlearning: A Parameter and Inversion-Based Approach

arXiv 2604.07386•183d45a5d6792abee33b42006cedc34c911872ee2002540817a64e768c691d9d
MCPMEVRAGRPSGTRUSTDESCblockchaincontact-networksdifferential-privacyepidemiologyfair-ordering','MEV-ACE'','post-quantum','AITH','ML-DSA','delegggenetic-algorithmgradient-attacklabel-leakagemachine-unlearningmodel-context-protocolmodel-inversionnode-level-DPpoisoning-attacksprompt-injectionretrieval-augmented-generationretriever-poisoningsupply-chain-securitysynthetic-datatool-poisoningtrusted-tool-descriptions

Paper metadata

arXiv ID
2604.07386
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
183d45a5d6792abee33b42006cedc34c911872ee2002540817a64e768c691d9d
Enrichment time
2026-04-10T07:23:35Z
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.

Label Leakage Attacks in Machine Unlearning: A Parameter and Inversion-Based Approach · Baitaphish