Pith. sign in

Title resolution pending

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

fields

cs.CR 1

years

2025 1

verdicts

CONDITIONAL 1

representative citing papers

Watermarking LLM-Generated Datasets in Downstream Tasks

cs.CR · 2025-06-16 · conditional · novelty 5.0

Watermarks such as trigger words, rare tokens, and grammatical style changes can often be detected in downstream models trained on LLM-generated data, but effectiveness and false positives vary sharply.

citing papers explorer

Showing 1 of 1 citing paper.

  • Watermarking LLM-Generated Datasets in Downstream Tasks cs.CR · 2025-06-16 · conditional · none · ref 2

    Watermarks such as trigger words, rare tokens, and grammatical style changes can often be detected in downstream models trained on LLM-generated data, but effectiveness and false positives vary sharply.