Kanishka's repositories
wugs-and-daxes
Collection of academic works in natural language processing, computational linguistics, and computational cognitive science that study the problem of novel words in models/agents.
inside-trumps-white-house
Code and analysis for authorship attribution of the op-ed in nytimes about the resistance in the president's administation
gpt2samples
R Package for exploring Samples of generated texts from Open AI's new GPT-2 language model
minicons-experiments
Code and analysis for the minicons paper
semantic-memory
Package to create a semantic-memory like repository of conceptual knowledge.
lm-induction
Code and analyses for the paper: "A Property Induction Framework for Neural Language Models" to be presented at CogSci 2022
minimal-category-abstraction
Minimal implementation of Kim and Smolensky (2021) along with bells and whistles introduced by Misra and Kim (2023)
cogsci2019
Code and Analysis for CogSci 2019 @ Montreal, Canada and ICCM 2019 @ Montreal, Canada
emnlp-bert-priming
Cleaned repository of our code and analysis of the paper "BERT as a Semantic Priming Subject: Exploring BERT's use of Lexical Cues to Inform Word Probabilities in Context"
fine-tuning-and-composition-in-transformers
This repo contains datasets and code for On the Interplay Between Fine-tuning and Composition in Transformers, by Lang Yu and Allyson Ettinger.
grounding-embeddings
Comparing human-based and machine-based meaning.
hypernymysuite
Hearst Patterns Revisited: Automatic Hypernym Detection from Large Text Corpora
inductive-reasoning-phenomena
A collection of phenomena observed in inductive reasoning experiments.
multiverse
A collection of useful utilities.
nafips2021
Code and analysis for my NAFIPS 2021 submission
pytorch-tutorials
Learning pytorch so that I can actually do NLP. Hiding code because impostor syndrome
rcpp-practice
Code for Practicing Rcpp to speed up R code :fire:
tattarrattat
Repository to study palindromic languages in neural networks
typicalityprobing
Code, Data, and Analyses for the probing LMs for typicality effects.