Research
My PhD thesis is on Data Selection for Generalization in Unimodal and Multimodal Models. How to select (data pruning), order (curriculum learning) and modify (data augmentation) training data to achieve better performance in deep learning models - are some of the questions I have explored in my work. I have also curated evaluation datasets for foundation models.
I am interested in understanding the role of training data and creating synthetic data for efficient learning in models.
|
Data Curation
|
Evaluating Very Long-Term Conversational Memory of LLM Agents
Adyasha Maharana, Dong-Ho Lee, Sergey Tulyakov, Mohit Bansal, Francesco Barbieri, Yuwei Fang
arXiv, 2024
arxiv /
website /
We present a dataset of very long conversations created in a semi-synthetic manner for evaluating long-range memory of LLMs. We find that long-context models struggle to retreieve and utilize relevant context.
Invited talk at Microsoft Turing
|
|
Exposing and Addressing Cross-Task Inconsistency in Unified Vision-Language Models
Adyasha Maharana, Amita Kamath, Christopher Clark, Mohit Bansal, Aniruddha Kembhavi
Transactions of Machine Learning Research (TMLR), 2024
arxiv /
data /
code /
website /
We curate a dataset for examining cross-task inconsistency in multi-task multimodal models and address this phenomena with a ranking-based loss objective.
|
|
D2 Pruning: Message Passing for Balancing Diversity & Difficulty in Data Pruning
Adyasha Maharana, Prateek Yadav, Mohit Bansal
ICLR, 2024
arxiv /
code /
We represent datasets as graphs to select the best-performing subset of a dataset in supervised, self-supervised and unsupervised pruning scenarios.
|
|
On curriculum learning for commonsense reasoning
Adyasha Maharana, Mohit Bansal
NAACL, 2022
paper /
code /
We demonstrate that paced curriculum learning improves in-domain and out-of-domain performance of models trained for commonsense reasoning.
|
|
Adversarial augmentation policy search for domain and cross-lingual generalization in reading comprehension
Adyasha Maharana, Mohit Bansal
Findings of EMNLP, 2020
arxiv /
code /
We show that mixing training data from a diverse set of adversarial augmentation strategies can improve in-domain, cross-domain as well as cross-domain research comprehension.
|
Multimodal Models
|
StoryDALL-E: Adapting Pretrained Text-to-Image Transformers for Story Continuation
Adyasha Maharana, Darryl Hannan, Mohit Bansal
European Conference on Computer Vision (ECCV), 2022
arxiv /
code /
demo /
We retro-fit open-source implementation of DALL-E to perform parallelized generation of a sequence of visual frames based on a given story. The generated images are conditioned on a source frame to improve generalization to unseen characters in the story.
|
|
Multimodal intent discovery from livestream videos
Adyasha Maharana, Quan Hung Tran, Franck Dernoncourt, Seunghyun Yoon, Trung Bui, Walter Chang, Mohit Bansal
Finding of NAACL, 2022
paper /
code /
We introduce methods for extracting intent from videos based on a curated dataset of software tutorial videos.
|
|
Improving generation and evaluation of visual stories via semantic consistency
Adyasha Maharana, Darryl Hannan, Mohit Bansal
NAACL, 2021
arxiv /
code /
We introduce Transformer-based models for performing story visualization by leveraging cyclic consistency with video captioning.
|
ML + Public Health
|
Detecting reports of unsafe foods in consumer product reviews
Adyasha Maharana, Kunlin Cai, Joseph Hellerstein, Yulin Hswen, Michael Munsell, Valentina Staneva, Miki Verma, Cynthia Vint, Derry Wijaya, Elaine O Nsoesie
Journal of American Medical Informatics Association (JAMIA) Open, 2019
paper /
We linked Amazon food product reviews to FDA food recalls from 2012 to 2014 and automate the detection of reports of unsafe food products.
|
|
Use of deep learning to examine the association of the built environment with prevalence of neighborhood adult obesity
Adyasha Maharana, Elaine Okanyene Nsoesie
Journal of American Medical Association (JAMA) Network Open, 2018
paper /
code /
We demonstrate the use of image embeddings for drawing correlation between built environment and neighborhood adult obesity rates. This work was covered in news and accompanied by an Editorial Comment.
Invited talk at ICLR SEDL Workshop 2021
|
|
Social media as a sentinel for disease surveillance: what does sociodemographic status have to do with it?
Elaine O Nsoesie, Luisa Flor, Jared Hawkins, Adyasha Maharana, Tobi Skotnes, Fatima Marinho, John S Brownstein
PLoS currents, 2016
paper /
code /
We apply machine-learning methods to explore spatial and temporal dengue event reporting trends on Twitter relative to confirmed cases in Brazil.
|
|