Pith. sign in

REVIEW 4 cited by

Three Bricks to Consolidate Watermarks for Large Language Models

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 2308.00113 v2 pith:6CLJOAMY submitted 2023-07-26 cs.CL cs.AI

classification cs.CLcs.AI
keywords generatedlanguagewatermarksdetectionlargemodelsnaturaltext
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
abstract

The task of discerning between generated and natural texts is increasingly challenging. In this context, watermarking emerges as a promising technique for ascribing generated text to a specific model. It alters the sampling generation process so as to leave an invisible trace in the generated output, facilitating later detection. This research consolidates watermarks for large language models based on three theoretical and empirical considerations. First, we introduce new statistical tests that offer robust theoretical guarantees which remain valid even at low false-positive rates (less than 10$^{\text{-6}}$). Second, we compare the effectiveness of watermarks using classical benchmarks in the field of natural language processing, gaining insights into their real-world applicability. Third, we develop advanced detection schemes for scenarios where access to the LLM is available, as well as multi-bit watermarking.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 4 Pith papers

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

  1. Beyond Heuristic Tuning: Power-Calibrated LLM Watermarking

    stat.ML 2026-07 accept novelty 7.0 of 10

    A power-calibrated statistical framework gives closed-form links from KGW watermark parameters (γ, δ) to detection power and KL distortion, enabling principled Pareto-optimal selection.

  2. Debiasing Watermarks for Large Language Models via Maximal Coupling

    stat.ML 2024-11 conditional novelty 6.0 of 10

    A new LLM watermarking scheme uses maximal coupling to embed a detectable signal while keeping the generated token distribution exactly unbiased.

  3. From Trade-off to Synergy: A Versatile Symbiotic Watermarking Framework for Large Language Models

    cs.CL 2025-05 reject novelty 5.0 of 10

    An entropy-driven fusion of logits-based and sampling-based watermarks is claimed to outperform existing LLM watermarking methods on all four evaluation axes.

  4. Temperature Matters: Enhancing Watermark Robustness Against Paraphrasing Attacks

    cs.CL 2025-06 reject novelty 4.0 of 10

    A watermark that seeds each token's sampling temperature from a hash of the previous h tokens is claimed to beat the Aaronson watermark under a 30% BERT paraphrase attack, based on a single ROC curve without error bars.

Pith tools