Papers by Katsiaryna Mirylenka

4 papers
Investigating Active Learning Sampling Strategies for Extreme Multi Label Text Classification (2022.lrec-1)

Copied to clipboard

Challenge: Large scale, multi-label text datasets with high numbers of different classes are expensive to annotate due to domain experts taking a lot of time working through all the classes.
Approach: They propose to build classifiers on multi-label text datasets using Active Learning to reduce labeling effort.
Outcome: The proposed classifiers can be used to reduce labeling effort on multi-label datasets.
Evaluating Pre-Trained Sentence-BERT with Class Embeddings in Active Learning for Multi-Label Text Classification (2022.aacl-short)

Copied to clipboard

Challenge: Existing Active Learning strategies for pre-trained transformer language models are limited and do not work well on domain-specific datasets.
Approach: They employ pre-trained transformer sentence embeddings to group samples with the same labels in the embeddable space on a legal document corpus.
Outcome: The proposed method performs significantly worse than baselines on two domain-specific datasets.
Reinforced Active Learning for Low-Resource, Domain-Specific, Multi-Label Text Classification (2023.findings-acl)

Copied to clipboard

Challenge: Modern text classification systems achieve excellent accuracy across tasks and corpora.
Approach: They propose a Reinforcement Learning policy that uses many different aspects of the data and task to select the most informative unlabeled subset dynamically over the course of the AL procedure.
Outcome: The proposed framework outperforms baselines on four complex multi-class, multi-label text classification datasets.
Adapting LLMs for Structured Natural Language API Integration (2024.emnlp-industry)

Copied to clipboard

Challenge: API integration is crucial for enterprise systems, but there are challenges in combining APIs based on user intent.
Approach: They propose a framework that leverages large language models to integrate APIs based on natural language input.
Outcome: The proposed framework improves performance over existing methods and RAGs based on open APIs . it can learn structural API constraints implicitly during training and retain structured knowledge .

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