Evaluating and explaining training strategies for zero-shot cross-lingual news sentiment analysis
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909330372034560 |
|---|---|
| author | Andrenšek, Luka Koloski, Boshko Pelicon, Andraž Lavrač, Nada Pollak, Senja Purver, Matthew |
| author_facet | Andrenšek, Luka Koloski, Boshko Pelicon, Andraž Lavrač, Nada Pollak, Senja Purver, Matthew |
| contents | We investigate zero-shot cross-lingual news sentiment detection, aiming to develop robust sentiment classifiers that can be deployed across multiple languages without target-language training data. We introduce novel evaluation datasets in several less-resourced languages, and experiment with a range of approaches including the use of machine translation; in-context learning with large language models; and various intermediate training regimes including a novel task objective, POA, that leverages paragraph-level information. Our results demonstrate significant improvements over the state of the art, with in-context learning generally giving the best performance, but with the novel POA approach giving a competitive alternative with much lower computational overhead. We also show that language similarity is not in itself sufficient for predicting the success of cross-lingual transfer, but that similarity in semantic content and structure can be equally important. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_20054 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Evaluating and explaining training strategies for zero-shot cross-lingual news sentiment analysis Andrenšek, Luka Koloski, Boshko Pelicon, Andraž Lavrač, Nada Pollak, Senja Purver, Matthew Computation and Language Artificial Intelligence We investigate zero-shot cross-lingual news sentiment detection, aiming to develop robust sentiment classifiers that can be deployed across multiple languages without target-language training data. We introduce novel evaluation datasets in several less-resourced languages, and experiment with a range of approaches including the use of machine translation; in-context learning with large language models; and various intermediate training regimes including a novel task objective, POA, that leverages paragraph-level information. Our results demonstrate significant improvements over the state of the art, with in-context learning generally giving the best performance, but with the novel POA approach giving a competitive alternative with much lower computational overhead. We also show that language similarity is not in itself sufficient for predicting the success of cross-lingual transfer, but that similarity in semantic content and structure can be equally important. |
| title | Evaluating and explaining training strategies for zero-shot cross-lingual news sentiment analysis |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2409.20054 |