Page: 1... 44 45 46 47 48 49 50 51 52
941 |
Does BERT Learn as Humans Perceive? Understanding Linguistic Styles through Lexica ...
|
|
|
|
Abstract:
Anthology paper link: https://aclanthology.org/2021.emnlp-main.510/ Abstract: People convey their intention and attitude through linguistic styles of the text that they write. In this study, we investigate lexicon usages across styles throughout two lenses: human perception and machine word importance, since words differ in the strength of the stylistic cues that they provide. To collect labels of human perception, we curate a new dataset, Hummingbird, on top of benchmarking style datasets. We have crowd workers highlight the representative words in the text that makes them think the text has the following styles: politeness, sentiment, offensiveness, and five emotion types. We then compare these human word labels with word importance derived from a popular fine-tuned style classifier like BERT. Our results show that the BERT often finds content words not relevant to the target style as important words used in style prediction, but humans do not perceive the same way even though for some styles (e.g., ...
|
|
Keyword:
Computational Linguistics; Language Models; Machine Learning; Machine Learning and Data Mining; Natural Language Processing
|
|
URL: https://dx.doi.org/10.48448/e4sm-9557 https://underline.io/lecture/37883-does-bert-learn-as-humans-perceivequestion-understanding-linguistic-styles-through-lexica
|
|
BASE
|
|
Hide details
|
|
942 |
Predicting emergent linguistic compositions through time: Syntactic frame extension via multimodal chaining ...
|
|
|
|
BASE
|
|
Show details
|
|
943 |
Just Ask! Evaluating Machine Translation by Asking and Answering Questions ...
|
|
|
|
BASE
|
|
Show details
|
|
944 |
BPM_MT: Enhanced Backchannel Prediction Model using Multi-Task Learning ...
|
|
|
|
BASE
|
|
Show details
|
|
945 |
An Analysis of Euclidean vs. Graph-Based Framing for Bilingual Lexicon Induction from Word Embedding Spaces ...
|
|
|
|
BASE
|
|
Show details
|
|
946 |
Statistically Significant Detection of Semantic Shifts using Contextual Word Embeddings ...
|
|
|
|
BASE
|
|
Show details
|
|
947 |
Lying Through One's Teeth: A Study on Verbal Leakage Cues ...
|
|
|
|
BASE
|
|
Show details
|
|
948 |
Iterative Paraphrastic Augmentation with Discriminative Span Alignment ...
|
|
|
|
BASE
|
|
Show details
|
|
949 |
Guilt by Association: Emotion Intensities in Lexical Representations ...
|
|
|
|
BASE
|
|
Show details
|
|
950 |
Findings of the WMT Shared Task on Machine Translation Using Terminologies ...
|
|
|
|
BASE
|
|
Show details
|
|
951 |
Knowledge-Interactive Network with Sentiment Polarity Intensity-Aware Multi-Task Learning for Emotion Recognition in Conversations ...
|
|
|
|
BASE
|
|
Show details
|
|
952 |
Does Vision-and-Language Pretraining Improve Lexical Grounding? ...
|
|
|
|
BASE
|
|
Show details
|
|
953 |
Translation Transformers Rediscover Inherent Data Domains ...
|
|
|
|
BASE
|
|
Show details
|
|
954 |
XLEnt: Mining a Large Cross-lingual Entity Dataset with Lexical-Semantic-Phonetic Word Alignment ...
|
|
|
|
BASE
|
|
Show details
|
|
956 |
GeneSis: A Generative Approach to Substitutes in Context ...
|
|
|
|
BASE
|
|
Show details
|
|
957 |
Avoiding Inference Heuristics in Few-shot Prompt-based Finetuning ...
|
|
|
|
BASE
|
|
Show details
|
|
958 |
Leveraging Word-Formation Knowledge for Chinese Word Sense Disambiguation ...
|
|
|
|
BASE
|
|
Show details
|
|
959 |
Generic resources are what you need: Style transfer tasks without task-specific parallel training data ...
|
|
|
|
BASE
|
|
Show details
|
|
960 |
Phrase-level Active Learning for Neural Machine Translation ...
|
|
|
|
BASE
|
|
Show details
|
|
Page: 1... 44 45 46 47 48 49 50 51 52
|
|