Publications
(*indicates equal contribution)
Much of my research is on the downstream applications of Keyphrase Generation and Stance Detection. Currently, I am interested in prompt engineering and leveraging Large Language Models (LLMs) for data augmentation purposes. I am also interested in explainability, robustness, and calibration of the deep learning models.
|
Stanceformer: Target-Aware Transformer for Stance Detection
Krishna Garg, Cornelia Caragea
ACL 2024 (Under Review)
Stanceformer provides superior performance, more accurate evaluation, and efficient utilization of compute power compared to GPT-3.5, Llama-2-7b-chat, Llama-2-13b-chat, and other SOTA models.
Preprint available by request :)
|
A New Direction in Stance Detection: Target-Stance Extraction in the Wild
Krishna Garg*, Yingjie Li*, Cornelia Caragea
ACL 2023 (Oral Presentation)
arXiv
/ code
|
Data Augmentation for Low-Resource Keyphrase Generation
Krishna Garg, Jishnu Ray Chowdhury, Cornelia Caragea
ACL 2023 (Findings) (Spotlight)
arXiv
/ code
|
Keyphrase Generation Beyond the Boundaries of Title and Abstract
Krishna Garg, Jishnu Ray Chowdhury, Cornelia Caragea
EMNLP 2022 (Findings)
arXiv
/ code
|
Teaching Experience
Teaching Assistant
CS 341: Programming Language Design and Implementation, Spring 2024, UIC
CS 341: Programming Language Design and Implementation, Fall 2023, UIC
CS 111: Programming Design I, Spring 2023, UIC
CS 251: Data Structures, Fall 2022, UIC
CS 480: Database Systems, Summers 2021, UIC
CS 362: Computer Design, Spring 2021, UIC
CS 341: Programming Language Design and Implementation, Spring 2020, UIC
CS 476: Programming Language Design, Fall 2019, UIC
CS 111: Program Design I, Legal and Public Policy Themes Section, Fall 2019, UIC
CS 341: Programming Language Design and Implementation, Fall 2018, UIC
|
[NLP]
Neural Machine Translation
NMT with RNN Models: (1) in Vanilla style, (2) with Sentencepiece, (3) using Pre-trained models from FairSeq
code
|
Disaster-Related Tweet Classification
BERT-based models for classification of tweets to facilitate disaster management
report
/ code
|
Travello
Extracting addresses from unstructured text (e.g.,restaurant blogs) using deep learning models like LSTM, BERT
report
/ code
|
Named Entity Recognition
PyTorch-based project to train a classifier to recognize entities like person, geographical entity, event, location, etc.
code
|
[Computer Vision & Machine Learning]
Visual Dialog: A Survey
Survey paper on recent advances in Visual Dialog task
report
|
Few Shot Learning
Ph.D. Qualifier report on Few-Shot Learning
report
|
optim-meta: Benchmarking meta-learning algorithms
Benchmarking optimization-based meta-learning algorithms
report
/ code
|
Hand Gesture Recognition
Using OpenCV2 library and deep learning-based methods
report
/ code
|
[Misc.]
Multi-path Routing Algorithm for Intra-domain Routing in Network Simulator-3
Bachelor Thesis with Prof. Virendra Singh Shekhawat at BITS Pilani
|
Achievements
Recipient of Peter and Deborah Wexler Graduate Scholarship ($5000), UIC
Recipient of GSC Travel Award ($500), UIC
Recipient of Merit Cum Need Scholarship (25% Fee Waiver), BITS Pilani
Ranked 6364/400k+, 5962/1.2m+, 13/135k+ in national entrance exams IIT-JEE, AIEEE, BITSAT
|
Invited Talks
2022 | DPI Research Scholars Workshop: Graduate School Prep and Planning
2023 | Keyphrase Generation: Redefining Data Science with NLP Applications | CS 418: Intro to Data Science
|
Service
Reviewer: ACL 2023, EMNLP 2023, NAACL 2024
Volunteer: ACL 2023, EMNLP 2022
Program Committee Member, EMNLP Industry Track 2023
Executive Committee Member, UIC Senate, 2023-24
Member, Committee on Budget, Planning, and Priorities, UIC, 2023-24
Representative of CS Dept in Graduate Student Council, UIC, 2022-23
|
|