Learning to Skim Text Transformer Replication

  • Fine-tuned an NLP model that surpassed previous state-of-the-art performance in text skimming tasks.
  • Achieved significant improvements over the proposed LSTM-jump model:
    • Increased accuracy by 5%
    • Doubled computational speed
  • Authored and presented a detailed technical report to the professor and class, demonstrating strong communication and presentation skills.

Technologies

  • Developed the project using a comprehensive tech stack:
    • Python
    • Google Colab
    • PyTorch
    • Scikit-learn
    • NumPy
    • Pandas
    • Hugging Face Transformers
    • SQL
    • Git
    • TensorFlow