20+ high-performance LLMs with recipes to pretrain, finetune and deploy at scale.

A family of open-sourced Mixture-of-Experts (MoE) Large Language Models

Tools for merging pretrained large language models.

Codes for the paper: "Continual Learning for Text Classification with Information Disentanglement Based Regularization"

Simple Conversational Data Augmentation for Semi-supervised Abstractive Conversation Summarization

Code and dataset fo paper: Understanding the Usage of Online Media for Parenting from Infancy to Preschool At Scale

Source codes for "Structure-Aware Abstractive Conversation Summarization via Discourse and Action Graphs"

Implementation of ICLR 2020 paper "Revisiting Self-Training for Neural Sequence Generation"

MediaSum: A Large-scale Media Interview Dataset for Dialogue Summarization

The source code for the Cutoff data augmentation approach proposed in this paper: "A Simple but Tough-to-Beat Data Augmentation Approach for Natural Language Understanding and Generation".

Source codes for the paper "Examining the Ordering of Rhetorical Strategies in Persuasive Requests"

Source codes for the paper "Local Additivity Based Data Augmentation for Semi-supervised NER"

Source codes for the paper "Multi-View Sequence-to-Sequence Models with Conversational Structure for Abstractive Dialogue Summarization"

"End-to-End Abstractive Summarization for Meetings" paper - Unofficial PyTorch Implementation

Knowledge Graph-Augmented Abstractive Summarization with Semantic-Driven Cloze Reward

PyTorch implementation of Contrastive Learning methods

Codebase for the Summary Loop paper at ACL2020

Language Model Baselines for PyTorch

annotated screenplays for 39 CSI:Crime Scene Investigation episodes for paper "Whodunnit? Crime Drama as a Case for Natural Language Understanding"

Beyond Accuracy: Behavioral Testing of NLP models with CheckList

Data and code associated with ICWSM 2020 paper about collective attention and descriptor phrases.

Code for WWW-20 Paper: HTML: Hierarchical Transformer-based Multi-task Learning for Volatility Prediction

"Let’s Make Your Request More Persuasive: Modeling Persuasive Strategies via Semi-Supervised Neural Nets on Crowdfunding Platforms"

MixText: Linguistically-Informed Interpolation of Hidden Space for Semi-Supervised Text Classification

Code for "Semi-Supervised Models via Data Augmentation for Classifying Interactive Affective Responses"