Muhammad Khalifa

xlm-roberta-ner. Named Entity Recognition with Pretrained XLM-RoBERTa

ThinkPRM. [TMLR] Process Reward Models That Think

GRACE. [EMNLP '23] Discriminator-Guided Chain-of-Thought Reasoning

fairseq-tagging. a Fairseq fork for sequence tagging/labeling tasks

PromptRank. [ACL 2023] Few-shot Reranking for Multi-hop QA via Language Model Prompting

intrinsic-source-citation. [COLM '24] Source-Aware Training Enables Knowledge Attribution in Language Models

icl-ensembling. [Me-FoMo ICLR 2023 - Oral] Exploring Demonstration Ensembling for In-context Learning

low-resource-seq-labeling. [EACL 2021] Self-training Pretrained LMs for Zero- and Few-shot Arabic Sequence Labeling

CoH-pytorch. pytorch implementation of the paper "Chain of Hindsight Aligns Language Models with Feedback"

Luhn-s-summarizer. An automatic text summarizer script based on Luhn's summarization algorithm

deep-learning-papers-notes. Thoughts and summaries of a collection deep-learning research papers.

transformer-diacritization. Arabic diacritization with Transformers

system-2-research. System 2 Reasoning Link Collection

fine-tunable. Keras Transfer Learning for NLP

nlp-useful-bash-commands. collection of useful bash commands for data manipulation in MLP

arabic-UlmFit. Python

utorrent-speed-limiter. Simple tool for windows to set and remove utorrent download speed limit based on whether user is idle or not.

keras-quora-question-pairs. A Keras model that addresses the Quora Question Pairs dyadic prediction task.