TA for Natural Language Processing

Serving as a teaching assistant for Professor Renfen Hu’s Natural Language Processing course.

My responsibilities included:

• Developing a concise manual on linear algebra and conducting four supplementary classes (totaling 12 hours) to introduce essential linear algebra concepts for NLP research. These sessions emphasized geometric intuition for students without prior background in the subject.

• Facilitating 4 tutorial sessions, demonstrating practical code implementation of common NLP tasks. I created and maintained a comprehensive tutorial repository showcasing model training and fine-tuning techniques using PyTorch and 🤗Transformers. This included implementations of models such as TextCNN, LSTM, BERT, and T5 for tasks like text classification and machine translation, as well as fine-tuning large language models such as Qwen for specialized applications like brain teasers.

• Enhancing and extending asynchronous API call code originally designed for OpenAI to support a wider range of models, including Claude and the LLaMA series, from various providers.