Papers by Anca Dinu

2 papers
It takes two to borrow: a donor and a recipient. Who’s who? (2024.findings-acl)

Copied to clipboard

Challenge: Existing methods for identifying the direction of borrowing are limited.
Approach: They propose strong benchmarks for automatic borrowing direction detection by using a borrowings dataset from the recent RoBoCoP database for five Romance languages.
Outcome: The proposed model improves the accuracy of the proposed task and proposes additional directions for future work.
RoBoCoP: A Comprehensive ROmance BOrrowing COgnate Package and Benchmark for Multilingual Cognate Identification (2023.emnlp-main)

Copied to clipboard

Challenge: Existing databases for romance cognates are scattered, incomplete, noisy, or have uncertain availability.
Approach: They propose to use etymological information to identify Romance cognates and borrowings from dictionaries to identify their ethymology.
Outcome: The proposed method achieves 94% accuracy on two pairs of Romance languages.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations