kushalarora

quantifying_exposure_bias

Accompanying repository for the paper: Why Exposure Bias Matters: An Imitation Learning Perspective of Error Accumulation in Language Generation

LMPL

Language Modeling as a Policy Learning library.

kushalarora.github.io

stable_entropy_hypothesis

transformers

🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.

al-folio

A beautiful, simple, clean, and responsive Jekyll theme for academics

allennlp

An open-source NLP research library, built on PyTorch.

allennlp-models

Officially supported AllenNLP models

awd-lstm-lm

LSTM and QRNN Language Model Toolkit for PyTorch

compound-pcfg

drex_rlhf

Doing D-Rex in RLHF for summarization

fairseq

Facebook AI Research Sequence-to-Sequence Toolkit written in Python.

fyrd

Submit functions and shell scripts to torque and slurm clusters or local machines using python.

GLUE-baselines

Repo for exploring multi-task learning approaches to learning sentence representations

kushalarora

Config files for my GitHub profile.

mauve

Package to compute Mauve, a similarity score between neural text and human text. Install with `pip install mauve-text`.

mgs

MLE-guided parameter search (https://arxiv.org/pdf/2006.03158.pdf)

OpenNMT-py

Open Source Neural Machine Translation in PyTorch

ParlAI

A framework for training and evaluating AI models on a variety of openly available dialogue datasets.

PoincareMaps

The need to understand cell developmental processes has spawned a plethora of computational methods for discovering hierarchies from scRNAseq data. However, existing techniques are based on Euclidean geometry which is not an optimal choice for modeling complex cell trajectories with multiple branches. To overcome this fundamental representation issue we propose Poincaré maps, a method harnessing the power of hyperbolic geometry into the realm of single-cell data analysis.

quant_exp_bias

Experiments to quantify exposure bias.

reinforce-joey

This is a fork of the awesome Joey-NMT with Reinforcement Learning algorithms like Policy Gradient, MRT and Advantage Actor Critic.

releasing-research-code

Tips for releasing research code in Machine Learning (with official NeurIPS 2020 recommendations)

seh_colm_rebuttal

text

Data loaders and abstractions for text and NLP

TreeNLG

A novel method of constrained decoding for neural NLG (NNLG) models

trl

Train transformer language models with reinforcement learning.

trlx

A repo for distributed training of language models with Reinforcement Learning via Human Feedback (RLHF)

visualbert

Code for the paper "VisualBERT: A Simple and Performant Baseline for Vision and Language"

wge

Workflow-Guided Exploration: sample-efficient RL agent for web tasks