1 |
EMBEDDIA tools output example corpus of Estonian, Croatian and Latvian news articles 1.0
|
|
|
|
BASE
|
|
Show details
|
|
2 |
Out of Thin Air: Is Zero-Shot Cross-Lingual Keyword Detection Better Than Unsupervised? ...
|
|
|
|
BASE
|
|
Show details
|
|
9 |
Keyword extraction datasets for Croatian, Estonian, Latvian and Russian 1.0
|
|
|
|
BASE
|
|
Show details
|
|
12 |
List of single-word male and female occupations in Slovenian
|
|
|
|
BASE
|
|
Show details
|
|
14 |
Slav-NER: the 3rd Cross-lingual Challenge on Recognition, Normalization, Classification, and Linking of Named Entities across Slavic languages ...
|
|
|
|
BASE
|
|
Show details
|
|
15 |
Slav-NER: the 3rd Cross-lingual Challenge on Recognition, Normalization, Classification, and Linking of Named Entities across Slavic languages ...
|
|
|
|
BASE
|
|
Show details
|
|
16 |
Evaluation of contextual embeddings on less-resourced languages ...
|
|
|
|
BASE
|
|
Show details
|
|
17 |
Simple Discovery of COVID IS WAR Metaphors Using Word Embeddings ...
|
|
|
|
BASE
|
|
Show details
|
|
18 |
Simple Discovery of COVID IS WAR Metaphors Using Word Embeddings ...
|
|
|
|
BASE
|
|
Show details
|
|
19 |
Investigating cross-lingual training for offensive language detection
|
|
|
|
In: PeerJ Comput Sci (2021)
|
|
Abstract:
Platforms that feature user-generated content (social media, online forums, newspaper comment sections etc.) have to detect and filter offensive speech within large, fast-changing datasets. While many automatic methods have been proposed and achieve good accuracies, most of these focus on the English language, and are hard to apply directly to languages in which few labeled datasets exist. Recent work has therefore investigated the use of cross-lingual transfer learning to solve this problem, training a model in a well-resourced language and transferring to a less-resourced target language; but performance has so far been significantly less impressive. In this paper, we investigate the reasons for this performance drop, via a systematic comparison of pre-trained models and intermediate training regimes on five different languages. We show that using a better pre-trained language model results in a large gain in overall performance and in zero-shot transfer, and that intermediate training on other languages is effective when little target-language data is available. We then use multiple analyses of classifier confidence and language model vocabulary to shed light on exactly where these gains come from and gain insight into the sources of the most typical mistakes.
|
|
Keyword:
Computational Linguistics
|
|
URL: https://doi.org/10.7717/peerj-cs.559 http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8237322/
|
|
BASE
|
|
Hide details
|
|
20 |
Temporal Integration of Text Transcripts and Acoustic Features for Alzheimer's Diagnosis Based on Spontaneous Speech
|
|
|
|
In: Front Aging Neurosci (2021)
|
|
BASE
|
|
Show details
|
|
|
|