Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Contrastive Learning of Sociopragmatic Meaning in Social Media

About

Recent progress in representation and contrastive learning in NLP has not widely considered the class of \textit{sociopragmatic meaning} (i.e., meaning in interaction within different language communities). To bridge this gap, we propose a novel framework for learning task-agnostic representations transferable to a wide range of sociopragmatic tasks (e.g., emotion, hate speech, humor, sarcasm). Our framework outperforms other contrastive learning frameworks for both in-domain and out-of-domain data, across both the general and few-shot settings. For example, compared to two popular pre-trained language models, our method obtains an improvement of $11.66$ average $F_1$ on $16$ datasets when fine-tuned on only $20$ training samples per dataset.Our code is available at: https://github.com/UBC-NLP/infodcl

Chiyu Zhang, Muhammad Abdul-Mageed, Ganesh Jawahar• 2022

Related benchmarks

TaskDatasetResultRank
Semantic Textual SimilaritySTS tasks (STS12, STS13, STS14, STS15, STS16, STS-B, SICK-R) various (test)
STS12 Score48.13
393
Transfer LearningSentEval Transfer tasks (test)
MR86.83
23
Crisis ClassificationCrisisOltea v1 (test)
Macro F196.01
14
Emotion DetectionEmoMoham v1 (test)
Macro F1 Score81.96
14
Hate Speech DetectionHateWas v1 (test)
Macro F157.65
14
Out-of-domain performance averageAverage Out-of-Domain
Macro F175.54
14
Twitter dataset performance averageAverage In-Domain
Macro F178.58
14
Sociopragmatic classification16 in-domain sociopragmatic datasets (test)
F1 (N=20)46.88
6
Sociopragmatic classification8 out-of-domain sociopragmatic datasets (test)
Avg F1 (N=20)42.19
6
Emotion ClassificationEmoBian Italian (test)
Macro F1 Score74.07
5
Showing 10 of 17 rows

Other info

Code

Follow for update