sustcsonglin Goto Github PK

stanford-beamer-presentation

This is an unofficial LaTeX Beamer presentation template for Stanford University.

state-spaces

Sequence Modeling with Structured State Spaces

storchastic

Stochastic Automatic Differentiation library for PyTorch.

streaming-llm

Efficient Streaming Language Models with Attention Sinks

struct-learning-with-flow

PyTorch Implementation of Unsupervised Learning of Syntactic Structure with Invertible Neural Projections

structuredlm_rtdt

The official code for paper "R2D2: Recursive Transformer based on Differentiable Tree for Interpretable Hierarchical Language Modeling".

strux

sustcsonglin.github.io

sustcsonglin_old.github.io

:page_facing_up: Elegant & friendly homepage (bio, tech portfolio, resume, doc...) template with Markdown and VuePress

t2fw

Fine-Tuning Pre-trained Transformers into Decaying Fast Weights

tcfft

tensor-fft

A implementation of an FFT algorithm targeting fp16 data to accelerate its processing by utilizing tensor cores

tn-lcfrs

Official Implementation of ACL2023: Unsupervised Discontinuous Constituency Parsing with Mildly Context-Sensitive Grammars

tn-pcfg

source code of NAACL2021 "PCFGs Can Do Better: Inducing Probabilistic Context-Free Grammars with Many Symbols“ and ACL2021 main conference "Neural Bilexicalized PCFG Induction"

torch-discounted-cumsum

Fast Discounted Cumulative Sums in PyTorch

training-kit

Open source on demand courses and cheat sheets for Git and GitHub

treebankpreprocessing

Python scripts preprocessing Penn Treebank and Chinese Treebank

triton

Development repository for the Triton language and compiler

triton-inside

Using triton to implement inside algorithms

udgn

Code for Unsupervised Dependency Graph Network paper

unsupconstparseeval

An Empirical Comparison of Unsupervised Constituency Parsing Methods

vuepress-homepage

zmsp

The Mingled Structured Predictor (Some tools for structured prediction, currently mainly for dependency parsing)

zoology

Understand and test language model architectures on synthetic tasks.