Writing
Papers, my dissertation, research positions and awards.
Publications
Transformer-based Models for Supervised Monocular Depth Estimation
Dissertation
Hallucination Span Detection in Large Language Models
- Benchmarked open-source LLMs against proprietary state-of-the-art models on hallucination span detection, finding them comparable in detection and subject to the same limitations.
- Showed open-source models are highly effective context retrievers, with end-to-end pipeline performance dropping by only 1–2% when proprietary retrieval engines were replaced.
- Explored prompt modification and temperature-varied ensembling of the best-performing model.
Research positions
Formal v/s Contextual Cues for Grammatical Gender Prediction
- Trained vector representations from the contextual and formal properties of words and compared their usability in a noun gender prediction task.
- Used gender-debiasing methods to isolate the gender information in a word vector, showing that the human reliance on semantic over phonological cues shows up in language models too.
Diachronic Distributed Word Representations as Models of Mental Lexicon
- Used a temporally sliced corpus to learn distributed word representations of child speech and child-directed speech under a curriculum-learning setting.
- Demonstrated the ease of usage and effectiveness of diachronic distributed word representations in modelling lexical development, presented at the ACL 2022 Student Research Workshop.
Awards
First Runner-Up, Global PyTorch Summer Hackathon 2020
Won with Rasoee against 2,500+ participants from 114 countries, and featured in PyTorch's official publication.
First Runner-Up, Microsoft CodeFunDo++ 2019
Won with ChainVoter, a blockchain-backed secure online voting application.
Research Week with Google 2022
One of 50 participants selected for the NLU track across undergraduate, master's and PhD students in India.
OpenCV Spatial AI Competition 2021
Among the top 210 finalists from over 1,500 submissions worldwide.