TrFP fingerprints LLMs from how representations change with depth
Fingerprints stay under 0.3 MiB and remain stable after fine-tuning, pruning, quantization, merging, and permutation.
Stories per day · 30 days
1 RSS
Relatedbatch-normalization1dnn1federated-learning1feed-forward-networks1llm-fingerprinting1model-fingerprinting1open-weights1provenance1spectral-analysis1watermarking1
3 stories, newest first · one card per story
Stories tagged #model-ownership
Training-free spectral fingerprints match related LLM checkpoints from weights
GATE singular-value features stay similar under instruction tuning, quantization, pruning, and distillation without prompts or forward passes.
Data-free attack removes batch-norm watermarks from federated models
Tests on two federated-learning schemes strip ownership marks without training data and with little accuracy loss.