Publications & Patents

Papers & patents

20+ papers spanning LLMs & recommendation systems (2025–present) and Dravidian-language NLP (2020–2022), plus 2 filed patents. Full list, links, and citation counts on Google Scholar.

Google Scholar profile ↗

Recent Publications · LLMs & Recommendation Systems

NAACL 2027 · forthcoming
LLM post-training for user profile generation for discovering sports segments at scale
EACL 2027 · forthcoming
Optimizing LLM augmentation for improving tail recommendations
GRAIL @ CIKM 2026 · forthcoming
From head to tail: optimizing LLM augmentations for improving sequential recommendations
GEM @ ACL 2026
Evaluating ASR quality at scale on TV entertainment platforms
RecSys 2025
Simulating discoverability for upcoming content on entertainment platforms
SIGIR 2025
Content moderation in TV search: balancing compliance, relevance, and user experience
Journal of Data, Information and Management · 2022
Multi-task learning in under-resourced Dravidian languages
COLING 2020 (PEOPLES Workshop)
KanCMD: Kannada codemixed dataset for sentiment analysis and offensive language detection

Earlier Publications · Dravidian-Language NLP (2020–2022)

Int'l Journal of Information Management Data Insights · 2022
Modeling homophobia and transphobia detection using data augmentation in a multilingual code-mixed setting
Advances in Computational Intelligence · 2022
Do images really do the talking? Analysing the significance of images in Tamil troll meme classification
Computer Speech & Language · 2022
Offensive language identification in low-resourced code-mixed Dravidian languages using pseudo-labeling
LT-EDI @ ACL 2022
The best of both worlds: dual channel language modeling for hope speech detection
DravidianLangTech @ ACL 2022 (TamilNLP)
DE-ABUSE@TamilNLP-ACL 2022: transliteration as data augmentation for abuse detection in Tamil
DravidianLangTech @ ACL 2022
Findings of the shared task on multi-task learning in Dravidian languages
Multigen @ ICON 2021
Hypers at ComMA@ICON: modelling aggressiveness, gender bias, and communal bias identification
PAKDD 2021
Domain identification of scientific articles using transfer learning and ensembles
ICCMC 2021
Evaluating pretrained transformer-based models for COVID-19 fake news detection
LoResMT @ EACL 2021
Attentive fine-tuning of transformers for translation of low-resourced languages
DravidianLangTech @ EACL 2021
IIITT@DravidianLangTech-EACL2021: transfer learning for offensive language detection
LT-EDI @ EACL 2021
IIITT@LT-EDI-EACL2021: hope speech detection
DravidianLangTech @ EACL 2021
UVCE-IIITT@DravidianLangTech-EACL2021: Tamil troll meme classification
CASE @ ACL 2021
IIITT at CASE 2021 Task 1: leveraging pretrained language models for multilingual protest detection

Sourced from an earlier academic portfolio and public paper listings — worth a quick check against Google Scholar for anything since revised. For citation counts, see Google Scholar.

Patents

Comcast · 2025
Systems for content filtering
Comcast · 2025
Systems for enhancing content discoverability in TV systems