Header

Search

DDIS goes to EMNLP!

Greetings from Budapest

DDIS was in Budapest for EMNLP 2026, where Daan, Selene, Noah, and Laurin presented the group’s work. Selene co-authored “Holmes-Graph: A Framework for Probing the Linguistic Competence of Graph-Based Text Embeddings,” which appeared in Findings of EMNLP. At the BlackboxNLP workshop, Noah and Laurin presented “Reproducing Transparent and Scrutable Recommendations: Exploring Open-Weight Models via Natural-Language User Profiles,” and Daan and Selene presented “Reproducing and Evaluating the Generalizability of Subliminal Learning in Open-Weight Models,” co-authored with Nathan. Stefan also attended, presenting “Probe Before you Answer! Answerability Probing for Abstention in Cross-Lingual Context-Based Question Answering” at the Conference on Machine Translation (WMT), together with collaborators outside the group.

Best Reproducibility Paper Award!

Congratulations to Stefan, whose paper “Probing Without Labels: Contrast-only Reproductions of Interpretability Results,” written with Peter Bloem, won the Best Reproducibility Paper Award at BlackboxNLP!

Subpages