Papers by Ana Uban
The Myth of Double-Blind Review Revisited: ACL vs. EMNLP (D19-1)
Copied to clipboard
| Challenge: | a double-blind review system enforces author anonymity during the review period . authors can be inferred with accuracy as high as 87% on ACL and 78% on EMNLP . |
| Approach: | They examine how well deep learning techniques can infer authors of a paper . authors found authors can be inferred with accuracy as high as 87% on ACL and 78% on EMNLP . |
| Outcome: | The proposed method can infer authors with 87% accuracy on ACL and 78% on EMNLP for the top 100 most prolific authors. |
Verba volant, scripta volant? Don’t worry! There are computational solutions for protoword reconstruction (2024.emnlp-main)
Copied to clipboard
Liviu Dinu, Ana Uban, Alina Cristea, Ioan-Bogdan Iordache, Teodor-George Marchitan, Simona Georgescu, Laurentiu Zoicas
| Challenge: | Existing methods for protoword reconstruction are limited to a few languages. |
| Approach: | They propose a new database of cognate words and etymons for the five main Romance languages and apply machine learning to it. |
| Outcome: | The proposed model achieves 90% accuracy in predicting protowords for Romance languages, surpassing state-of-the-art models and features. |
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
Liviu Dinu, Ana Uban, Alina Cristea, Anca Dinu, Ioan-Bogdan Iordache, Simona Georgescu, Laurentiu Zoicas
| 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. |