UMD at EMNLP
November 12, 2024

Terrapins in Florida on language models and human languages.
November 12-16, Empirical Methods in Natural Language Processing (EMNLP) and the Conference on Computational Natural Language Learning (CoNLL) are co-located in Miami, and feature three presentations of Terrapin work, from Rosa, Sathvik, Joselyn, Leslie, Katherine and Allison, undergraduate collaborators Kamala Sreepada and Robert Hopkins, and Naomi Feldman.
EMNLP
- Eun-kyong Rosa Lee, Sathvik Nair and Naomi Feldman / A psycholinguistic evaluation of language models' sensitivity to argument roles
CoNNL
- Joselyn Rodriguez, Kamala Sreepada, Ruolan Leslie Famularo, Sharon Goldwater and Naomi Feldman / Self-supervised speech representations display some human-like cross-linguistic perceptual abilities
- Katherine Howitt, Sathvik Nair, Allison Dods and Robert Melvin Hopkins / Generalizations across filer-gap dependencies in neural language models