Layerwise Convergence Fingerprints for Runtime Misbehavior Detection in Large Language Models

Documenti analoghi

• CORVUS: Red-Teaming Hallucination Detectors via Internal Signal Camouflage in Large Language Models
  di: Min, Nay Myat, et al.
  Pubblicazione: (2026)
• Unified Neural Backdoor Removal with Only Few Clean Samples through Unlearning and Relearning
  di: Min, Nay Myat, et al.
  Pubblicazione: (2024)
• FNF: Functional Network Fingerprint for Large Language Models
  di: Liu, Yiheng, et al.
  Pubblicazione: (2026)
• REEF: Representation Encoding Fingerprints for Large Language Models
  di: Zhang, Jie, et al.
  Pubblicazione: (2024)
• SoK: Large Language Model Copyright Auditing via Fingerprinting
  di: Shao, Shuo, et al.
  Pubblicazione: (2025)