Pith. sign in

REVIEW 2 cited by

Debiasing Watermarks for Large Language Models via Maximal Coupling

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2411.11203 v2 pith:MPKMKJTZ submitted 2024-11-17 stat.ML cs.CLcs.CRcs.LGstat.ME

classification stat.MLcs.CLcs.CRcs.LGstat.ME
keywords textgreenlanguagemodelsqualitywatermarkingapproachbias
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Watermarking language models is essential for distinguishing between human and machine-generated text and thus maintaining the integrity and trustworthiness of digital communication. We present a novel green/red list watermarking approach that partitions the token set into ``green'' and ``red'' lists, subtly increasing the generation probability for green tokens. To correct token distribution bias, our method employs maximal coupling, using a uniform coin flip to decide whether to apply bias correction, with the result embedded as a pseudorandom watermark signal. Theoretical analysis confirms this approach's unbiased nature and robust detection capabilities. Experimental results show that it outperforms prior techniques by preserving text quality while maintaining high detectability, and it demonstrates resilience to targeted modifications aimed at improving text quality. This research provides a promising watermarking solution for language models, balancing effective detection with minimal impact on text quality.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Optimal Estimation of Watermark Proportions in Hybrid AI-Human Texts

    stat.ML 2025-06 conditional novelty 7.0 of 10

    For continuous-score text watermarks, the proportion of watermarked tokens in mixed AI-human text is identifiable and can be estimated at the minimax-optimal rate.

  2. Watermark in the Classroom: A Conformal Framework for Adaptive AI Usage Detection

    stat.AP 2025-07 conditional novelty 6.0 of 10

    Standard, hierarchical, and weighted conformal prediction applied to LLM watermark scores can control false-positive rates when detecting guideline-violating AI edits in simulated classroom essays.

Pith tools