Apr 2026	Recipient of the Fonds de recherche du Québec Master’s Research (FRQNT) Scholarship for the 2026–2027 year.	Award
Apr 2026	BRIDGE: Predicting Human Task Completion Time From Model Performance accepted at ICML 2026.	Paper
Feb 2026	New preprint: BRIDGE: Predicting Human Task Completion Time From Model Performance.	Preprint
Aug 2025	LLMs Can Compensate for Deficiencies in Visual Representations accepted at EMNLP 2025.	Paper
Jun 2025	New preprint: LLMs Can Compensate for Deficiencies in Visual Representations.	Preprint
Jan 2025	MMTEB: Massive Multilingual Text Embedding Benchmark accepted at ICLR 2025.	Paper
Aug 2024	Gave a talk on in-context learning capabilities of LLMs for MT (slides) at the SNLP Reading Group, Microsoft Research India.	Talk
Aug 2024	RomanSETU received the 🏆 Senior Area Chair Award at ACL 2024.	Award
May 2024	RomanSETU, ICL study for MT and Data Pruning for MT accepted at ACL 2024.	Paper
May 2024	Leverage Class-Specific Accuracy to Guide Data Generation for Improving Image Classification accepted at ICML 2024.	Paper
Mar 2024	New preprint: On the low-shot transferability of [V]-Mamba.	Preprint
Jan 2024	New preprint: ICL abilities in LLMs for MT.	Preprint
Jan 2024	New preprint: Airavata, an instruction-tuned Hindi LLM (code).	Preprint
Nov 2023	IndicTrans2 accepted at TMLR.	Paper
Nov 2023	Presented tutorial on Developing SOTA MNMT Systems for Related Languages at AACL-IJCNLP 2023.	Talk
May 2023	New preprint: IndicTrans2, the first open-source model supporting all 22 Scheduled Indian languages (code).	Preprint
Jan 2023	A Federated Approach for Hate Speech Detection accepted at EACL 2023 (code).	Paper