2 |
An Information-Theoretic Characterization of Morphological Fusion ...
|
|
|
|
BASE
|
|
Show details
|
|
3 |
Learning Data Augmentation Schedules for Natural Language Processing ...
|
|
|
|
BASE
|
|
Show details
|
|
4 |
A Simple Geometric Method for Cross-Lingual Linguistic Transformations with Pre-trained Autoencoders ...
|
|
|
|
BASE
|
|
Show details
|
|
5 |
Exploring Pre-Trained Transformers and Bilingual Transfer Learning for Arabic Coreference Resolution ...
|
|
|
|
BASE
|
|
Show details
|
|
6 |
Evaluating the Morphosyntactic Well-formedness of Generated Texts ...
|
|
|
|
BASE
|
|
Show details
|
|
7 |
I Wish I Would Have Loved This One, But I Didn't -- A Multilingual Dataset for Counterfactual Detection in Product Review ...
|
|
|
|
BASE
|
|
Show details
|
|
8 |
MultiEURLEX - A multi-lingual and multi-label legal document classification dataset for zero-shot cross-lingual transfer ...
|
|
|
|
BASE
|
|
Show details
|
|
9 |
IR like a SIR: Sense-enhanced Information Retrieval for Multiple Languages ...
|
|
|
|
BASE
|
|
Show details
|
|
10 |
Chinese Opinion Role Labeling with Corpus Translation: A Pivot Study ...
|
|
|
|
BASE
|
|
Show details
|
|
11 |
Mitigating Language-Dependent Ethnic Bias in BERT ...
|
|
|
|
Abstract:
Anthology paper link: https://aclanthology.org/2021.emnlp-main.42/ Abstract: BERT and other large-scale language models (LMs) contain gender and racial bias. They also exhibit other dimensions of social bias, most of which have not been studied in-depth, and some of which vary depending on the language. In this paper, we study ethnic bias and how it varies across languages by analyzing and mitigating ethnic bias in monolingual BERT for English, German, Spanish, Korean, Turkish, and Chinese. To observe and quantify ethnic bias, we develop a novel metric called Categorical Bias score. Then we pro- pose two methods for mitigation; first using a multilingual model, and second using contextual word alignment of two monolingual models. We compare our proposed methods with monolingual BERT and show that these methods effectively alleviate the ethnic bias. Which of the two methods works better de- pends on the amount of NLP resources available for that language. We additionally experiment with Arabic and Greek to ...
|
|
Keyword:
Data Management System; Gender Studies; Machine translation; Natural Language Processing; Neural Network
|
|
URL: https://dx.doi.org/10.48448/0s0e-9493 https://underline.io/lecture/37807-mitigating-language-dependent-ethnic-bias-in-bert
|
|
BASE
|
|
Hide details
|
|
12 |
Genre as Weak Supervision for Cross-lingual Dependency Parsing ...
|
|
|
|
BASE
|
|
Show details
|
|
13 |
Synthetic Data Augmentation for Zero-Shot Cross-Lingual Question Answering ...
|
|
|
|
BASE
|
|
Show details
|
|
14 |
Visually Grounded Reasoning across Languages and Cultures ...
|
|
|
|
BASE
|
|
Show details
|
|
15 |
Students Who Study Together Learn Better: On the Importance of Collective Knowledge Distillation for Domain Transfer in Fact Verification ...
|
|
|
|
BASE
|
|
Show details
|
|
16 |
On the Relation between Syntactic Divergence and Zero-Shot Performance ...
|
|
|
|
BASE
|
|
Show details
|
|
|
|