I am a Symbolic Systems master’s student at Stanford University. I am currently doing NLP research with the Georgia Tech Financial Services Innovation Lab, supervised by Dr. Sudheer Chava (Alton M. Costley Chair Professor of Finance, Georgia Tech). I am also exploring cognitive mechanisms of temporal attention with Dr. Jamshed Bharucha (Distinguished Fellow, Dartmouth College).
During my undergraduate studies, I have been fortunate to be advised by Dr. Ram Prasad Krishnamoorthy (former Marie Curie Fellow and Associate Dean, Sai University) and Dr. Krithi Ramamritham (Vijay and Sita Vashee Chair Professor of Computer Science, IIT Bombay).
My research in the United States was made possible in part by the Global Undergraduate Exchange Program. I was one of the five students in India to be awarded a scholarship to pursue one semester of fully-funded study in Computer Science in the United States.
Email / CV / Google Scholar / GitHub
Michael Galarnyk*, Siddharth Lohani*, Vidhyakshaya Kannan, Sagnik Nandi, Aman Patel, Liqin Ye, Arnav Hiray, Rutwik Routu, Prasun Banerjee, Siddhartha Somani, Sudheer Chava
KDD 2026 · Datasets and Benchmarks Track · pp. 8882–8893
Nikita Tatarinov, Vidhyakshaya Kannan*, Haricharana Srinivasa*, Arnav Raj, Harpreet Singh Anand, Varun Singh, Aditya Luthra, Ravij Lade, Agam Shah, Sudheer Chava
ACL 2026 · Main Conference · Oral · pp. 3323–3359
paper / ACL Anthology / code
See Research for all publications. * Equal contribution.
My work spans long-context language model evaluation, knowledge graphs, retrieval-augmented generation, and multimodal financial documents. I am also interested in cognitive mechanisms of temporal attention.
* Equal contribution.
Michael Galarnyk*, Siddharth Lohani*, Vidhyakshaya Kannan, Sagnik Nandi, Aman Patel, Liqin Ye, Arnav Hiray, Rutwik Routu, Prasun Banerjee, Siddhartha Somani, Sudheer Chava
KDD 2026 · Datasets and Benchmarks Track · pp. 8882–8893
A toolkit and dataset for parsing long IPO filings into standardized sections and analyzing their text, tables, and images.
Nikita Tatarinov, Vidhyakshaya Kannan*, Haricharana Srinivasa*, Arnav Raj, Harpreet Singh Anand, Varun Singh, Aditya Luthra, Ravij Lade, Agam Shah, Sudheer Chava
ACL 2026 · Main Conference · Oral · pp. 3323–3359
paper / ACL Anthology / code
A knowledge-graph framework for evaluating long-context reasoning across multi-hop retrieval, set operations, and answer plurality, using 20,139 question–answer pairs and 16 LLMs.
Michael Galarnyk*, Rutwik Routu*, Vidhyakshaya Kannan*, Kosha Bheda, Prasun Banerjee, Agam Shah, Sudheer Chava
EMNLP 2025 · System Demonstrations · pp. 667–677
A retrieval-augmented assistant for conference checklist responses, evaluated with a curated dataset of 1,975 ACL checklist responses.
Vignesh Ethiraj*, Ashwath D*, Sidhanth Menon*, Divya Vijay*, Vidhyakshaya Kannan*
EMNLP 2025 · Industry Track · pp. 2449–2460
A telecom-specific embedding model adapted with triplet-loss fine-tuning for domain-specific semantic retrieval.
Course Assistant · Stanford University
Autumn 2026 · Instructor: Dr. Monica Lam
Course assistant for Stanford’s course on building AI agents.
Teaching Assistant · The Gen Academy
Instructors: Aishwarya Srinivasan (ex Google, Microsoft, IBM) and Arvind Narayanamurthy (ex Microsoft, Adobe, IBM)
Supported learners working on agentic AI projects. Cohort date ranges are from Maven.
Sai University
Supported more than 120 students across four semesters through peer tutoring, weekly office hours, and problem sets.