Code and resources for the paper: "Neural Reverse Engineering of Stripped Binaries using Augmented Control Flow Graphs"

DIRTY: Augmenting Decompiler Output with Learned Variable Names and Types

The official code for PRIMERA: Pyramid-based Masked Sentence Pre-training for Multi-document Summarization

SacreROUGE is a library dedicated to the use and development of text generation evaluation metrics with an emphasis on summarization.

A set of utilities for running few-shot prompting experiments on large-language models

Utilities used by the Deep Program Understanding team

Code for the paper: "On the Bottleneck of Graph Neural Networks and Its Practical Implications"

Language Models of Code are Few-Shot Commonsense Learners (EMNLP 2022)

Mapping Language to Code in a Programmatic Context

PyTorch code for the RetoMaton paper: "Neuro-Symbolic Language Modeling with Automaton-augmented Retrieval" (ICML 2022)

The official repository for Efficient Long-Text Understanding Using Short-Text Models (Ivgi et al., 2022) paper

The official code of EMNLP 2022, "SCROLLS: Standardized CompaRison Over Long Language Sequences".

Code for "Typilus: Neural Type Hints" PLDI 2020

Hoppity

Repo for ICML23 "Why do Nearest Neighbor Language Models Work?"

Code for the paper "On the Expressivity Role of LayerNorm in Transformers' Attention" (Findings of ACL'2023)

A highly sophisticated sequence-to-sequence model for code generation

Code for the paper "A Structural Model for Contextual Code Changes"

System to count the people entering and leaving an entrance, using a DNN as a detector (YOLO) and a tracking algorithm to count and track (CSRT)

cppminer produces a code2seq compatible datasets from C++ code bases.

JEMMA: An Extensible Java dataset for Many ML4Code Applications

An implementation of model parallel autoregressive transformers on GPUs, based on the DeepSpeed library.

Code for the paper "Symmetric Machine Theory of Mind", presented at ICML 2022.

Single-Node Attack For Fooling Graph Neural Networks

Repo for ICLR22 "Capturing Structural Locality in Non-parametric Language Models"

A Fast, Extensible Trainer and Extensions for Pytorch

Wrapper to do text generation with VHellendoorn's PolyCoder model

Github Crawler

Code for the model presented in the paper: "code2seq: Generating Sequences from Structured Representations of Code"