The benchmark for LLMs designed to tackle one of the most knowledge-intensive tasks in data science: writing feature engineering code, which requires domain knowledge in addition to a deep understa…

🥇Samsung AI Challenge 2021 1등 솔루션입니다🥇

This repository contains a demonstrative implementation for pooling-based models, e.g., DeepPyramidion complementing our paper "Sparsifying Transformer Models with Trainable Representation Pooling."

Implementation of 🦩 Flamingo, state-of-the-art few-shot visual question answering attention net out of Deepmind, in Pytorch

OCR Annotations from Amazon Textract for Industry Documents Library

JSON Schema format for storing datasets details, documents processed contents, and documents annotations in the document understanding domain.

The evaluator covering all of the metrics required by tasks within the DUE Benchmark.

The code related to the baselines from NeurIPS 2021 paper "DUE: End-to-End Document Understanding Benchmark."

docTR (Document Text Recognition) - a seamless, high-performing & accessible library for OCR-related tasks powered by Deep Learning.

A curated list of resources for Document Understanding (DU) topic

A modular framework for vision & language multimodal research from Facebook AI Research (FAIR)

Fully featured implementation of Routing Transformer

Beautiful visualizations of how language differs among document types.

A fast and highly accurate differentiable Top-k operator from the "Successive Halving Top-k Operator" AAAI'21 paper.

The repository contains model implementations and data described in the paper: From Dataset Recycling to Multi-Property Extraction and Beyond.

Pytorch library for fast transformer implementations

Data and additional information regarding the paper: Contract Discovery. Dataset and a Few-Shot Semantic Retrieval Challenge with Competitive Baselines (to appear in Findings of EMNLP).

A tiny scalar-valued autograd engine and a neural net library on top of it with PyTorch-like API

A curated list of pretrained sentence and word embedding models

Inference of high-quality embeddings of complex networks into the hyperbolic disk.

Statistical Rethinking with PyTorch and Pyro