Study finds LLM 'fingerprints' can be contagious across models
A new arXiv paper reports that the fingerprint by which a classifier identifies the LLM behind a text is not fixed but can be contagious, passing between models.
Read the full story at arXiv cs.CL ↗
Timeline · 1 report
- 2026-11-16 05:00 · arXiv cs.CL
Contagious Fingerprints: When LLMs Inherit Another Model's Identity