Every input given to the generator and every text it produced during full-scale evaluations of the Thicket watermark against published watermarking schemes. All schemes were run with the same model, prompts, sampling settings and seeds, so each row of one config has an exact counterpart in every other config of the same generator (same promptid, same sample). The dataset lets readers inspect what went into the model and what came out, compare schemes on identical inputs (quality, diversity, length) and reuse the texts for robustness or detection experiments. Each config is one stage (split type) and each of its splits is one scheme (the names in the tables below), so all splits of a config…
Independent publisher
Aditya Pujari
RootAccess4Life
Models in Library0
Datasets in Library1
Models on Hugging Face7
Followers1
Compare the published evidence
This profile connects the 1 model and dataset record shown below to their upstream cards and sources. The library counts describe SAVRN's coverage, rather than this publisher's entire output.
Published tasks: Text generation.
Reported licenses: cc-by-4.0. Open a record to review its specific license and conditions.
Compare model specifications, source dates, and hardware estimates on the individual records. Publisher identity verification is separate from a model's quality, deployment readiness, or suitability for your work.