activity
20172022
most citedSynchromesh: Reliable code generation from pre-trained language models

66 citations · 98 across the 4 of their papers we have counts for

collaborators

5 papers

cs.SE202230 cited

Repairing Bugs in Python Assignments Using Large Language Models

Jialu Zhang, José Cambronero, Sumit Gulwani +4

Students often make mistakes on their introductory programming assignments as part of their learning process. Unfortunately, providing custom repairs for these mistakes can require…

cs.LG202266 cited

Synchromesh: Reliable code generation from pre-trained language models

Gabriel Poesia, Oleksandr Polozov, Vu Le +4

Large pre-trained language models have been used to generate code,providing a flexible interface for synthesizing programs from natural language specifications. However, they often…

cs.AI2021

Multi-modal Program Inference: a Marriage of Pre-trainedLanguage Models and Component-based Synthesis

Kia Rahmani, Mohammad Raza, Sumit Gulwani +5

Multi-modal program synthesis refers to the task of synthesizing programs (code) from their specification given in different forms, such as a combination of natural language and ex…

cs.SE2018

Learning Quick Fixes from Code Repositories

Reudismam Rolim, Gustavo Soares, Rohit Gheyi +2

Code analyzers such as Error Prone and FindBugs detect code patterns symptomatic of bugs, performance issues, or bad style. These tools express patterns as quick fixes that detect…

cs.HC20172 cited

TraceDiff: Debugging Unexpected Code Behavior Using Trace Divergences

Ryo Suzuki, Gustavo Soares, Andrew Head +5

Recent advances in program synthesis offer means to automatically debug student submissions and generate personalized feedback in massive programming classrooms. When automatically…