Name: Aaron Mueller
Type: User
Company: Northeastern ≡ The Technion
Bio: NLP ∩ Robustness ∩ Interpretability ∩ Multilinguality
Twitter: amuuueller
Location: Boston, MA ≡ Haifa, Israel
Blog: aaronmueller.github.io
Aaron Mueller's Projects
Aaron Mueller's personal website.
Syntactic evaluation sets, attribute-varying grammars, and code for replicating the CLAMS paper. ACL 2020.
A python package to setup topic classification fine-tuning, run contextualized topic modeling, and run TCCTMs
Adapting the Don't Stop Pretraining approach for multilingual applications. Modified by Aaron Mueller and Nathaniel Weir.
Config files for easy setup on new UNIX-based machines
Earley parser implementation.
Code for "How to Plant Trees in Language Models" (ACL 2023).
Basic pipeline for running different sized GPT models and plotting the results
Few-shot evaluation of language models. Fork for the BabyLM competition (CoNLL '23).
Investigation of different methods of multilingual fine-tuning for document classification with mBERT.
Trying out finite-state transducers.
Utility for analyzing Transformer based representations of language.
Basic IBM-style machine translation models with various decoding methods.
Multilingual causal mediation analysis
Generating stories given prompts using GPT-2. We also try diverse decoding!
nshell: a basic shell environment written in C
Implementation of Hierarchical Recurrent Encoder-Decoder (HRED) model for narrative generation in ParlAI.
Hidden Markov Model tagger
Implementing smoothed n-gram language models.
Using sparse coding to find distributed representations used by neural networks.
Code and data for In-context Learning Generalizes, But Not Always Robustly: The Case of Syntax
Code for the paper "Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer"
A PyTorch framework for creating, running, and reproducing experiments on seq2seq models.
For foreign editions of Wiktionary, extract derivations on each page (if they exist).