Language models frequently generate outputs in unintended languages or scripts, a phenomenon known as off-target generation. While existing research has focused on language selection, the dimension of script knowledge remains understudied: before any linguistic understanding can occur, users must recognize the graphic...
David Kletz, Sandra Mitrovic, Ljiljana Dolamic et al.· Findings of the Association...· 0 citations
A pilot application will enable a more efficient interaction of clinical practitioners with the documents related to the medical history of a patient, and enable a Question/Answer interaction paradigm.
L. Ruinelli, Amos Colombo, O. Lithgow-Serrano et al.· Swiss Text Analytics Confere...· 0 citations
This paper investigates how script knowledge is distributed across the layers of LLMs using two complementary interpretability methods: logistic regression probing and logit-lens analysis, which show that script commitment consistently occurs at the very last layers of the LLMs.
David Kletz, Sandra Mitrovic, Itay Sabato et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.