SAFESEAL is a key-conditioned LLM watermarking framework using tournament sampling for synonym substitution and a contrastive detector that reports 98.2% detection, 0.983 BERTScore, and 0.963 entity similarity while claiming robustness to attacks.
Stealing part of a production language model
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Robust LLM Watermarking with Minimal Semantic Distortion for IP Protection
SAFESEAL is a key-conditioned LLM watermarking framework using tournament sampling for synonym substitution and a contrastive detector that reports 98.2% detection, 0.983 BERTScore, and 0.963 entity similarity while claiming robustness to attacks.