RAFT: Rationale adaptor for few-shot abusive language detection

by   Punyajoy Saha, et al.
IIT Kharagpur

Abusive language is a concerning problem in online social media. Past research on detecting abusive language covers different platforms, languages, demographies, etc. However, models trained using these datasets do not perform well in cross-domain evaluation settings. To overcome this, a common strategy is to use a few samples from the target domain to train models to get better performance in that domain (cross-domain few-shot training). However, this might cause the models to overfit the artefacts of those samples. A compelling solution could be to guide the models toward rationales, i.e., spans of text that justify the text's label. This method has been found to improve model performance in the in-domain setting across various NLP tasks. In this paper, we propose RAFT (Rationale Adaptor for Few-shoT classification) for abusive language detection. We first build a multitask learning setup to jointly learn rationales, targets, and labels, and find a significant improvement of 6 F1 on the rationale detection task over training solely rationale classifiers. We introduce two rationale-integrated BERT-based architectures (the RAFT models) and evaluate our systems over five different abusive language datasets, finding that in the few-shot classification setting, RAFT-based models outperform baseline models by about 7 competitively to models finetuned on other source domains. Furthermore, RAFT-based models outperform LIME/SHAP-based approaches in terms of plausibility and are close in performance in terms of faithfulness.


page 1

page 2

page 3

page 4


How to Solve Few-Shot Abusive Content Detection Using the Data We Actually Have

Due to the broad range of social media platforms and their user groups, ...

Abusive and Threatening Language Detection in Urdu using Boosting based and BERT based models: A Comparative Approach

Online hatred is a growing concern on many social media platforms. To ad...

A New cross-domain strategy based XAI models for fake news detection

In this study, we presented a four-level cross-domain strategy for fake ...

Self-Supervision Meta-Learning for One-Shot Unsupervised Cross-Domain Detection

Deep detection models have largely demonstrated to be extremely powerful...

UNIDECOR: A Unified Deception Corpus for Cross-Corpus Deception Detection

Verbal deception has been studied in psychology, forensics, and computat...

Cross-Platform and Cross-Domain Abusive Language Detection with Supervised Contrastive Learning

The prevalence of abusive language on different online platforms has bee...

Examining the Limitations of Computational Rumor Detection Models Trained on Static Datasets

A crucial aspect of a rumor detection model is its ability to generalize...

Please sign up or login with your details

Forgot password? Click here to reset