Dipankar Srirag
Ph.D. Candidate · UNSW Sydney
I am a Ph.D. candidate at the UNSW-NLP group at University of New South Wales, advised by Dr. Aditya Joshi, Dr. Padmanesan Narasimhan, and Dr. Salil Kanhere, and funded by the NHMRC Ideas grant.
My Ph.D. project develops multilingual conversational agents for emergency triage, grounded in real electronic health records. I am currently exploring simultaneous machine translation for dialects of Arabic as a core component of this work.
My broader research background is in dialect robustness: I have built benchmarks and parameter-efficient adaptation methods that evaluate and close the performance gap NLP models exhibit on non-standard English varieties, focusing on sentiment, sarcasm, slang, and conversation understanding across Australian, Indian, and British English.
News
- Jun 2025
I started my Ph.D. at the School of Computer Science and Engineering, UNSW Sydney.
- May 2025
Our paper, BESSTIE: a benchmark for sentiment and sarcasm classification across varieties of English, is accepted to the Findings of ACL: ACL 2025. Find the dataset on 🤗 Hugging Face.
- Apr 2025
I presented our work on dialect adapters for decoder models at the main proceedings of NAACL 2025 in Albuquerque.
- Jan 2025
Our paper, Evaluating Dialect Robustness of Language Models via Conversation Understanding is published at the SUMEval Workshop @ COLING 2025.
- May 2024
I started working as a Research Assistant at the School of Computer Science and Engineering, UNSW Sydney.
Selected Publications
BibTeX
@misc{srirag-etal-2026-triagesim,
title = "{TriageSim}: A Conversational Emergency Triage Simulation Framework from Structured Electronic Health Records",
author = "Srirag, Dipankar and Nguyen, Quoc Dung and Joshi, Aditya and Narasimhan, Padmanesan and Kanhere, Salil",
year = "2026",
eprint = "2603.10035",
archivePrefix = "arXiv",
primaryClass = "cs.CL",
url = "https://arxiv.org/abs/2603.10035"
}
Copied!BibTeX
@misc{singh-etal-2025-nekminit,
title = "Nek Minit: Harnessing Pragmatic Metacognitive Prompting for Explainable Sarcasm Detection of {A}ustralian and {I}ndian {E}nglish",
author = "Singh, Ishmanbir and Srirag, Dipankar and Joshi, Aditya",
year = "2025",
eprint = "2505.15095",
archivePrefix = "arXiv",
primaryClass = "cs.CL",
url = "https://arxiv.org/abs/2505.15095"
}
Copied!BibTeX
@inproceedings{srirag-etal-2025-besstie,
title = "{BESSTIE}: A Benchmark for Sentiment and Sarcasm Classification for Varieties of {E}nglish",
author = "Srirag, Dipankar and Joshi, Aditya and Painter, Jordan and Kanojia, Diptesh",
booktitle = "Findings of the Association for Computational Linguistics: ACL 2025",
month = jul,
year = "2025",
address = "Vienna, Austria",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.findings-acl.441/",
doi = "10.18653/v1/2025.findings-acl.441",
pages = "8413--8429"
}
Copied!BibTeX
@inproceedings{srirag-etal-2025-predicting,
title = "Predicting the Target Word of Game-playing Conversations using a Low-Rank Dialect Adapter for Decoder Models",
author = "Srirag, Dipankar and Joshi, Aditya and Eisenstein, Jacob",
booktitle = "Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 2: Short Papers)",
month = apr,
year = "2025",
address = "Albuquerque, New Mexico",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.naacl-short.2/",
doi = "10.18653/v1/2025.naacl-short.2",
pages = "8--17"
}
Copied!