Publications

Drug Safety Agents Using Graphs and Ontologies

AAAI Workshop on Health Intelligence / 2026

Can LLMs support drug-safety review while staying grounded in biomedical evidence and reducing hallucination risk?

View paper

Cite

BibTeX

@article{Mathialagan2026.02.04.26345582,
  author       = {Mathialagan, Clint Solomon and Nip, Alexander and Bhat, Ashwani},
  title        = {Drug Safety Agents Using Graphs and Ontologies},
  elocation-id = {2026.02.04.26345582},
  year         = {2026},
  doi          = {10.64898/2026.02.04.26345582},
  publisher    = {Cold Spring Harbor Laboratory Press},
  URL          = {https://www.medrxiv.org/content/early/2026/02/05/2026.02.04.26345582},
  eprint       = {https://www.medrxiv.org/content/early/2026/02/05/2026.02.04.26345582.full.pdf},
  journal      = {medRxiv}
}

Self-Aware Feedback-Based Self-Learning in Large-Scale Conversational AI

NAACL / 2022

When an AI system learns from user feedback, how does it avoid learning from the consequences of its own past mistakes?

View paper

Cite

BibTeX

@inproceedings{ponnusamy-etal-2022-self,
  title     = {Self-Aware Feedback-Based Self-Learning in Large-Scale Conversational {AI}},
  author    = {Ponnusamy, Pragaash and Mathialagan, Clint Solomon and Aguilar, Gustavo and Ma, Chengyuan and Guo, Chenlei},
  editor    = {Loukina, Anastassia and Gangadharaiah, Rashmi and Min, Bonan},
  booktitle = {Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies: Industry Track},
  month     = jul,
  year      = {2022},
  address   = {Hybrid: Seattle, Washington + Online},
  publisher = {Association for Computational Linguistics},
  url       = {https://aclanthology.org/2022.naacl-industry.36/},
  doi       = {10.18653/v1/2022.naacl-industry.36},
  pages     = {324--333}
}

PersonalTM: Transformer Memory for Personalized Retrieval

SIGIR / 2023

Can a retrieval system remember enough about a user to become more useful, while still remaining scalable as information changes?

View paper

Cite

BibTeX

@inproceedings{Lian_2023,
  series     = {SIGIR '23},
  title      = {PersonalTM: Transformer Memory for Personalized Retrieval},
  url        = {http://dx.doi.org/10.1145/3539618.3592037},
  DOI        = {10.1145/3539618.3592037},
  booktitle  = {Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval},
  publisher  = {ACM},
  author     = {Lian, Ruixue and Lu, Sixing and Solomon, Clint and Aguilar, Gustavo and Ponnusamy, Pragaash and Han, Jialong and Ma, Chengyuan and Guo, Chenlei},
  year       = {2023},
  month      = july,
  pages      = {2256--2260},
  collection = {SIGIR '23}
}

A Vocabulary-Free Multilingual Neural Tokenizer for End-to-End Task Learning

RepL4NLP at ACL 2022 / 2022

What if tokenization could be learned with the task, instead of being frozen before the model ever understands the language?

View paper

Cite

BibTeX

@inproceedings{mofijul-islam-etal-2022-vocabulary,
  title     = {A Vocabulary-Free Multilingual Neural Tokenizer for End-to-End Task Learning},
  author    = {Mofijul Islam, Md and Aguilar, Gustavo and Ponnusamy, Pragaash and Solomon Mathialagan, Clint and Ma, Chengyuan and Guo, Chenlei},
  editor    = {Gella, Spandana and He, He and Majumder, Bodhisattwa Prasad and Can, Burcu and Giunchiglia, Eleonora and Cahyawijaya, Samuel and Min, Sewon and Mozes, Maximilian and Li, Xiang Lorraine and Augenstein, Isabelle and Rogers, Anna and Cho, Kyunghyun and Grefenstette, Edward and Rimell, Laura and Dyer, Chris},
  booktitle = {Proceedings of the 7th Workshop on Representation Learning for NLP},
  month     = may,
  year      = {2022},
  address   = {Dublin, Ireland},
  publisher = {Association for Computational Linguistics},
  url       = {https://aclanthology.org/2022.repl4nlp-1.10/},
  doi       = {10.18653/v1/2022.repl4nlp-1.10},
  pages     = {91--99}
}

Personalized Query Rewriting in Conversational AI Agents

arXiv / 2020

Can conversational assistants better understand what users mean by learning from their own interaction history?

View paper

Cite

BibTeX

@misc{roshanghias2020personalizedqueryrewritingconversational,
  title         = {Personalized Query Rewriting in Conversational AI Agents},
  author        = {Alireza Roshan-Ghias and Clint Solomon Mathialagan and Pragaash Ponnusamy and Lambert Mathias and Chenlei Guo},
  year          = {2020},
  eprint        = {2011.04748},
  archivePrefix = {arXiv},
  primaryClass  = {cs.AI},
  url           = {https://arxiv.org/abs/2011.04748}
}

VIP: Finding Important People in Images

CVPR / 2015

In a group photo, can a computer infer who matters most-not just who is visible?

View paper

Cite

BibTeX

@InProceedings{Mathialagan_2015_CVPR,
  author    = {Solomon Mathialagan, Clint and Gallagher, Andrew C. and Batra, Dhruv},
  title     = {VIP: Finding Important People in Images},
  booktitle = {Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
  month     = {June},
  year      = {2015}
}

CloudCV: Large-Scale Distributed Computer Vision as a Cloud Service

Mobile Cloud Visual Media Computing / 2015

What if advanced computer vision tools were accessible through the cloud instead of locked behind specialized local infrastructure?

View paper

Cite

BibTeX

@inbook{Agrawal_2015,
  title     = {CloudCV: Large-Scale Distributed Computer Vision as a Cloud Service},
  ISBN      = {9783319247021},
  url       = {http://dx.doi.org/10.1007/978-3-319-24702-1_11},
  DOI       = {10.1007/978-3-319-24702-1_11},
  booktitle = {Mobile Cloud Visual Media Computing},
  publisher = {Springer International Publishing},
  author    = {Agrawal, Harsh and Mathialagan, Clint Solomon and Goyal, Yash and Chavali, Neelima and Banik, Prakriti and Mohapatra, Akrit and Osman, Ahmed and Batra, Dhruv},
  year      = {2015},
  pages     = {265--290}
}