talks

2023

• Developing SOTA MNMT Systems for Related Languages (Slides)
  Tutorial at AACL-IJCNLP 2023

• QLoRA: Efficient Finetuning of Quantized LLMs (Paper)
  ML Collective NLP Reading Group

• Few-Shot Parameter-Efficient Fine-Tuning is Better and Cheaper than In-Context Learning (Paper)
  ML Collective NLP Reading Group

• Sparks of Artificial General Intelligence: Early experiments with GPT-4 (Paper)
  ML Collective NLP Reading Group

• Mixture of Soft Prompts for Controllable Data Generation (Paper, Annotated Paper)
  ML Collective NLP Reading Group

• Does Localization Inform Editing? Surprising Differences in Causality-Based Localization vs. Knowledge Editing in Language Models (Paper, Slides)
  Cohere For AI NLP Reading Group

• Attention with Linear Biases Enables Input Length Extrapolation (Paper, Annotated Paper, Video)
  ML Collective NLP Reading Group

• Training a Language Model on a Single GPU in One Day (Paper, Annotated Paper, Video)
  ML Collective NLP Reading Group

2022

• Scaling Instruction-Finetuned Language Models (Flan-PaLM) (Paper, Slides)
  ML Collective NLP Reading Group

• Knowledge Neurons in Pretrained Transformers Review (Paper, Slides)
  ML Collective NLP Reading Group

• Learning from Mistakes based on Class Weighting with Application to Neural Architecture Search (Paper, Slides)
  PXie Lab

2021

• Facts as Experts: Adaptable and Interpretable Neural Memory over Symbolic Knowledge Review (Paper, Slides)
  ML Collective NLP Reading Group, Unicode Research

• Research Seminar for Undergraduates (Slides, Video)
  Unicode Research

• Deep Probabilistic Programming Review (Paper, Slides, Video)
  Unicode Research

• Auto-encoding Variational Bayes Review (Paper, Slides, Video)
  Unicode Research