Papers by Sumegh Roychowdhury

6 papers
Data-Efficient Methods For Improving Hate Speech Detection (2023.findings-eacl)

Copied to clipboard

Challenge: Existing methods for hate speech detection are data-hungry and require large datasets.
Approach: They propose an input-level data augmentation technique EasyMix to improve hate speech detection in english and multilingual datasets.
Outcome: The proposed method improves the performance across english and multilingual datasets by 1% and 2-8%.
IIT-KGP at COIN 2019: Using pre-trained Language Models for modeling Machine Comprehension (D19-60)

Copied to clipboard

Challenge: Using pre-trained language models, we can model machine comprehension using commonsense reasoning.
Approach: They propose a machine comprehension model that leverages pre-trained language models over commonsense knowledge bases.
Outcome: The proposed model improves on baseline models and other commonsense knowledge bases.
CRUSH: Contextually Regularized and User anchored Self-supervised Hate speech Detection (2022.findings-naacl)

Copied to clipboard

Challenge: Recent advances in NLP have often been used to mitigate the spread of hate speech and cyber-bullying on social networks.
Approach: They propose a framework for hate speech detection using user-anchored self-supervision and contextual regularization to learn better representations of hateful content.
Outcome: The proposed approach secures 1-12% improvement in test set metrics over best performing approaches on two types of tasks and multiple popular English language social networking datasets.
Representation Learning for Conversational Data using Discourse Mutual Information Maximization (2022.naacl-main)

Copied to clipboard

Challenge: Existing language modeling pretraining objectives do not take structural information of conversational text into account.
Approach: They propose a structure-aware Mutual Information based loss-function DMI for training dialog-representation models that captures the inherent uncertainty in response prediction.
Outcome: The proposed model outperforms strong baseline models on nine diverse tasks.
Generative or Discriminative? Revisiting Text Classification in the Era of Transformers (2025.emnlp-main)

Copied to clipboard

Challenge: generative classifiers exhibit lower sample complexity but higher asymptotic error in simple linear settings, a trade-off that remains unexplored in the transformer era.
Approach: They propose to evaluate generative and discriminative architectures for text classification using a generative model that learns the conditional probability distribution P (y|x) generative models are known to work better in low-data settings, giving rise to the classical 'two regimes' phenomenon for classification.
Outcome: The proposed models show that the classical 'two regimes' manifests distinctly across different architectures and training paradigms.
Exploring Ordinality in Text Classification: A Comparative Study of Explicit and Implicit Techniques (2024.findings-acl)

Copied to clipboard

Challenge: Ordinal classification (OC) is a key task in natural language processing with applications in various domains such as sentiment analysis, rating prediction, and more.
Approach: They propose to tackle ordinal classification (OC) through the implicit semantics of the labels . they propose to use a classical explicit approach and an implicit approach that organically engages the semantics.
Outcome: The proposed methods are based on pre-trained language models and offer strategic recommendations based upon specific settings.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations