Divyansha Lachi
Machine Learning Ph.D. Student @ University of Pennsylvania
Email: divyansha1115@gmail.com
About Me
Hi! I'm Divyansha, a Ph.D. student in Machine Learning at the University of Pennsylvania, Institute of Technology, where I am fortunate enough to be advised by Prof. Eva Dyer. My main areas of interest are graph machine learning and neuro-inspired AI. My current research focuses on developing scalable frameworks for multi-graph pre-training and new methods for representation learning, particularly in domains with complex and unstructured data that challenge conventional approaches. I’m passionate about figuring out how the brain works and what makes it tick. I believe that by understanding the brain, we can unlock new insights for science and AI. If you are interested in collaborating, feel free to reach out!
News
GraphFM accepted to TMLR
Excited to share that to share that GraphFM has been accepted to the Transactions on Machine Learning Research (TMLR)!
Read the paperLOG 2025 Acceptance 🎉
Excited to share that to share that RGP, my work from an internship at SAP, has been accepted to Learning on Graphs (LOG) 2025 conference!
Read the paperNeurIPS 2025 Acceptances 🎉
Excited to share that NuCLR has been accepted to the main NeurIPS 2025 conference, and our works RELATE and BTS have been accepted to the New Perspectives in Graph Machine Learning workshop!
Joining the University of Pennsylvania
I’ve transferred to the University of Pennsylvania, continuing my Ph.D. under the guidance of Prof. Eva Dyer.
Internship at SAP
I’m interning at SAP with the Business Foundation Model team. I’m based at the SAP office in Bellevue, WA, USA.
Genomic bottleneck 🧬 published at PNAS
Our paper on the genomic bottleneck has been published in the Proceedings of the National Academy of Sciences (PNAS).
Read the paperStochastic Genomic Bottleneck 🧠 preprint released
We’ve released a preprint of our work on the Stochastic Genomic Bottleneck, which will be presented at NAISys 2024.
View preprintGraphFM 🌐 preprint released
We've released a preprint for our work on GraphFM, a new framework for multi-graph pretraining.
View preprintStarted Ph.D. at Georgia Tech
I've begun my journey as a Machine Learning Ph.D. student at Georgia Institute of Technology.
Latest Publication
GraphFM: A generalist graph transformer that learns transferable representations across diverse domains
Transactions on Machine Learning Research (TMLR), 2025
Education
Ph.D. in Machine Learning
Georgia Institute of Technology
Advised by Prof. Eva Dyer
Ph.D. in Machine Learning
Georgia Institute of Technology
Advised by Prof. Eva Dyer
B.Tech. in Computer Science and Engineering
National Institute of Technology Silchar
Intermediate Science (Physics, Chemistry, Mathematics, Biology)
Rukmani Birla Modern High School, Jaipur
Experience
Graduate Research Assistant
University of Pennsylvania
Advised by Prof. Eva Dyer
Research Scientist Intern
SAP
Advised by Dr. Tom Palczewski
Graduate Research Assistant
Georgia Institute of Technology
Advised by Prof. Eva Dyer
Research Assistant
Cold Spring Harbor Lab
Advised by Prof. Anthony Zador
Research Intern
Brown University
Advised by Prof. Thomas Serre
Research Intern
Max Plank Institute for Brain Research
Advised by Prof. Moritz Helmstaedter
Research Intern
International Institute of Information Technology Hyderabad
Advised by Prof. Suryakanth V Gangashetty