Thibaud Gloaguen

Watch your steps: Dormant Adversarial Behaviors that Activate upon LLM Finetuning

LLM Fingerprinting via Semantically Conditioned Watermarks

Fewer Weights, More Problems: A Practical Attack on LLM Pruning

Watermarking Diffusion Language Models

Black-Box Detection of Language Model Watermarks

Discovering Spoofing Attempts on Language Model Watermarks

Discovering Clues of Spoofed LM Watermarks