This is the official repository of Team Anastasia, which achieved 1st place in Subtask 1 of the SemEval 2025 challenge.

Implement named entity recognition (NER) using regex and fine-tuned LLM, with a total of 15 categories. The ultimate goal is to apply the model to detect personally identifiable information (PII) in student writing.

(1) Train large language models to help people with automatic essay scoring. (2) Extract essay features and train new tokenizer to build tree models for score prediction.

Use LLMs to answer difficult science questions

Code for the paper "ELLEN: Extremely Lightly Supervised Learning For Efficient Named Entity Recognition" (LREC-COLING 2024)

Predict which Tweets are about real disasters and which ones are not Microsoft DeBERTa

This project was developed for a Kaggle competition focused on detecting Personally Identifiable Information (PII) in student writing. The primary objective was to build a robust model capable of identifying PII with high recall. The DeBERTa v3 transformer model was chosen for this task after comparing its performance with other transformer models.