For one-shot token watermarking with a min-entropy constraint, the authors derive a closed-form max-min detection rate and show that a correlated channel with balanced random partitions achieves it.
Truthfulqa: Measuring how models mimic human falsehoods
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Optimized Couplings for Watermarking Large Language Models
For one-shot token watermarking with a min-entropy constraint, the authors derive a closed-form max-min detection rate and show that a correlated channel with balanced random partitions achieves it.