Study shows LoRA fine-tunes leak training-data domain via embedding-space fingerprints
A new arXiv paper presents an embedding-space fingerprinting method that can recover the training-data domain of a LoRA fine-tuned model from the model artifact alone.
Read the full story at arXiv cs.AI ↗
Timeline · 1 report
- 2026-11-06 05:00 · arXiv cs.AI
Embedding-Space Fingerprinting of LoRA Fine-Tunes: Recovering the Training-Data Domain from the Model Artifact Alone