1 |
NB-MLM: Efficient Domain Adaptation of Masked Language Models for Sentiment Analysis ...
|
|
|
|
Abstract:
Anthology paper link: https://aclanthology.org/2021.emnlp-main.717/ Abstract: While Masked Language Models (MLM) are pre-trained on massive datasets, the additional training with the MLM objective on domain or task-specific data before fine-tuning for the final task is known to improve the final performance. This is usually referred to as the domain or task adaptation step. However, unlike the initial pre-training, this step is performed for each domain or task individually and is still rather slow, requiring several GPU days compared to several GPU hours required for the final task fine-tuning. We argue that the standard MLM objective leads to inefficiency when it is used for the adaptation step because it mostly learns to predict the most frequent words, which are not necessarily related to a final task. We propose a technique for more efficient adaptation that focuses on predicting words with large weights of the Naive Bayes classifier trained for the task at hand, which are likely more relevant than the ...
|
|
Keyword:
Computational Linguistics; Language Models; Machine Learning; Machine Learning and Data Mining; Natural Language Processing
|
|
URL: https://dx.doi.org/10.48448/bg4j-x794 https://underline.io/lecture/37952-nb-mlm-efficient-domain-adaptation-of-masked-language-models-for-sentiment-analysis
|
|
BASE
|
|
Hide details
|
|
2 |
A Comparative Study of Lexical Substitution Approaches based on Neural Language Models ...
|
|
|
|
BASE
|
|
Show details
|
|
3 |
HHMM at SemEval-2019 Task 2: Unsupervised frame induction using contextualized word embeddings
|
|
|
|
BASE
|
|
Show details
|
|
5 |
RUSSE'2018: A Shared Task on Word Sense Induction for the Russian Language ...
|
|
|
|
BASE
|
|
Show details
|
|
6 |
How much does a word weigh? Weighting word embeddings for word sense induction ...
|
|
|
|
BASE
|
|
Show details
|
|
7 |
RUSSE'2018 : a shared task on word sense induction for the Russian language
|
|
|
|
BASE
|
|
Show details
|
|
8 |
Negative Sampling Improves Hypernymy Extraction Based on Projection Learning ...
|
|
|
|
BASE
|
|
Show details
|
|
9 |
Negative Sampling Improves Hypernymy Extraction Based on Projection Learning ...
|
|
|
|
BASE
|
|
Show details
|
|
11 |
Human and Machine Judgements for Russian Semantic Relatedness ...
|
|
|
|
BASE
|
|
Show details
|
|
12 |
Negative sampling improves hypernymy extraction based on projection learning
|
|
|
|
BASE
|
|
Show details
|
|
13 |
Human and machine judgements for Russian semantic relatedness
|
|
|
|
BASE
|
|
Show details
|
|
14 |
Human And Machine Judgements For Russian Semantic Relatedness ...
|
|
|
|
BASE
|
|
Show details
|
|
|
|