SHIRAN DUDY

I'm a postdoc at CU Boulder

Hi. Welcome! I Recently joined CU Boulder as postdoc working on developing classroom assistants in educational setting. I work at the intersection of RL, NLP, and dialogue systems. I care for the ethical and human-centered systems as well. You can see here my papers, repo projects, and the research I've been working on. Hope you enjoy ;-)

Outdoor Study Group

PROFESSIONAL BIO

Current Research

My main Focus is on developing bots to facilitate discussion and learning in a classroom setting. Aside from that I continue to work on questions around bias, and aspects of personalization following my thesis.

PUBLICATIONS

Recent Research Papers

noun_personalization_1575827.png

2021

S. Dudy, S. Bedrick, B. Webber

Empirical Methods in Natural Language Processing (EMNLP)

type_coverage.png

The figure shows the percentage of Types (%) predicted stratified by frequency in known SotA language models 

2020

Won 2nd place award 

2020

S. Dudy, S. Bedrick

Eval4NLP Workshop

Empirical Methods in Natural Language Processing (EMNLP)

poster, code, clip

block2.png

2018

S. Dudy, S. Bedrick

DeepLo Workshop

Association for Computational Linguistics (ACL)

poster, code

process.jpg

2018

S. Dudy, S. Bedrick, M. Asgari, A. Kain

Computer Speech and Langauge (CSL)

fst_edited.jpg

2018

S. Dudy, S. Xu, S. Bedrick, D.Smith

SCLeM Workshop

Association for Computational Linguistics (NAACL)

poster, code

types_coverage_ny200.png

The figure shows the percentage of Types (%) predicted stratified by frequency in a categorical vs. continuous language models 

2020

S. Dudy, S. Bedrick

WiNLP Workshop

Association for Computational Linguistics (ACL)

noise.png

2019

R. Dong, D. Smith, S. Dudy, S. Bedrick

SPLAT Workshop

North American Association for Computational Linguistics (NAACL)

GIT REPOS

From a few projects I've worked on

GitHub-Mark-120px-plus.png

FST SPECIALIZER

2017

Python library to create Special Fsts as described in OpenFst source code. This specializer library enables sigma, rho, and phi operations over fsts over using the command line operations of OpenFst. To have a small footprint, this functionality was implemented in Cython and included generating C++ files.

Moby-logo1.png

DOCKER IMAGES FOR FST TOOLS

2016

Openfst and Pywrapfst library images that provide an openfst with pywrapfst environment on ubuntu 14.4. The most recent images are the pynini_1.4.1 and the fst_1.6.1 and their installation details as well as content is available.

GitHub-Mark-120px-plus.png

OBFUSCATION DETECTION WITH DEEP LEARNING

2016

The basic code for obfuscation detection provides a trained classifier to analyze and detect the encoding language used to obfuscate parts of code in order to hide there sensitive information or potnetially harmful behavior.

GitHub-Mark-120px-plus.png

PROSODY DETECTION

2015

Code for different sequence classifiers as described in the paper “Textual Prediction of Prosodic Prominence in Spontaneous Speech with Sequence Classifiers”. There are viterbi, greedy first and second order and, local search first second and third order.