English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

There are concerns that the ability of language models (LMs) to generate highquality synthetic text can be misused to launch spam, disinformation, orpropaganda. Therefore, the research community is actively working on developingapproaches to detect whether a given text is organic or synthetic. While thisis a useful first step, it is important to be able to further fingerprint theauthor LM to attribute its origin. Prior work on fingerprinting LMs is limitedto attributing synthetic text generated by a handful (usually < 10) ofpre-trained LMs. However, LMs such as GPT2 are commonly fine-tuned in a myriadof ways (e.g., on a domain-specific text corpus) before being used to generatesynthetic text. It is challenging to fingerprinting fine-tuned LMs because theuniverse of fine-tuned LMs is much larger in realistic scenarios. To addressthis challenge, we study the problem of large-scale fingerprinting offine-tuned LMs in the wild. Using a real-world dataset of synthetic textgenerated by 108 different fine-tuned LMs, we conduct comprehensive experimentsto demonstrate the limitations of existing fingerprinting approaches. Ourresults show that fine-tuning itself is the most effective in attributing thesynthetic text generated by fine-tuned LMs.

Fingerprinting Fine-tuned Language Models in the Wild