Home » Research

Category Archives: Research

Subscribe

Archives

Categories

Attribution-NonCommercial-ShareAlike 4.0 International

Except where otherwise noted, content on this site is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International license.

NYU GSTEM student visits during the summer of 2025

Ayla Zhang will join our research group this summer through the NYU GSTEM program. NYU GSTEM is a summer program for high school juniors that allows them to participate in research laboratories. The NYU Courant Institute of Mathematical Sciences offers the program and helps promote STEM to traditionally underrepresented groups, particularly females and minorities. Ayla will be working on a programming language project as part of our funded NSF project on combating technical debt in Machine Learning systemsbroader impacts.

(more…)

Slides and poster for FASE ’25 tool paper on imperative Deep Learning refactoring now available

The slides and our poster for our FASE ’25 formal tool demonstration paper on our work on automated refactoring of imperative Deep Learning programs to graph execution are now available!

(more…)

Tool demo paper on refactoring imperative DL programs to graph execution accepted at FASE ’25

Our paper entitled, “Hybridize functions: A tool for automatically refactoring imperative Deep Learning programs to graph execution,” has been accepted at the 2025 International Conference on Fundamental Approaches to Software Engineering (FASE) as a formal tool demonstration! Out of 31 papers (including 4 tool papers), 11 (including 1 tool paper) were accepted, amounting to a 35% acceptance rate. Congrats to Tatiana, Mehdi, Nan, and Anita!

Best paper award nomination at AISafety ’24

I am happy to announce that our paper entitled, “ReLESS: A framework for assessing safety in Deep Learning systems,” has been nominated for a best paper award at AI Safety ’24! Congrats to Nan and Anita. It’s Nan’s first workshop paper!

Paper on reliably refactoring Deep Learning systems accepted at AISafety ’24

Our paper on reliability refactoring Deep Learning systems has been accepted to the 2024 AISafety workshop at the International Joint Conference on Artificial Intelligence (IJCAI ’24). Congratulations to Nan and Anita!

Multiple fully-funded Ph.D. student positions in combating technical debt in Machine Learning (ML) systems in New York City

I am currently seeking multiple fully-funded Ph.D. students interested in programming languages and software engineering research for an NSF-funded project on combating technical debt in Machine Learning (ML) systems. The project—based in the heart of New York City—focuses on facilitating the long-lived evolution of ML systems through automated refactoring.

Potential research topics explored during the project may include (static/dynamic) program and data analysis and transformation, empirical software engineering, natural language processing (NLP), and Large Language Models (LLMs). Successful candidates will be expected to work on projects that generally yield open-source developer tool research prototypes, plug-ins to popular IDEs, build systems, or static analyzers. Applicants may find additional information on the PI’s web page and should also apply to the City University of New York (CUNY) Graduate Center (GC) Ph.D. program in Computer Science (deadline January 15) following a discussion with the PI. Students wishing to start earlier should speak with the PI.

(more…)

Received three-year NSF research grant on combating technical debt in Machine Learning systems as PI

I am pleased to announce that I have received a three-year standard research grant from the National Science Foundation (NSF) Software & Hardware Foundations (SHF) program as principal investigator (PI) for a project entitled “Knowledge, Methodologies, and Tool-support for Combating Technical Debt in Machine Learning Systems.” The total grant amount is ~$600K.

(more…)

Tatiana presenting on DL refactoring at ASE ’23

Tatiana presenting at ASE ’23

Slides for ASE ’23 NIER paper on imperative Deep Learning refactoring now available

Slides for our ASE ’23 NIER paper on our ongoing work towards automated refactoring of imperative Deep Learning programs to graph execution are now available. The talk will take place tomorrow at 1:54 pm CEST.

(more…)

Preprint of ASE ’23 DL refactoring paper now available

A preprint of our ASE ’23 paper on refactoring imperative Deep Learning programs to graphs is now available.