Lectures

Course Overview (Jan 8)

Content

  • Course logistics
  • What is natural language processing?
  • What are the features of natural language?
  • What do we want to do with NLP?
  • What makes it hard?

Slides

Course Oveview

Reading Material

Text Classification (Jan 10)

Content

  • Defining features
  • Building a rule-based classifier
  • Training a logistic regression based classifier
  • Evaluating classification

Slides

Text Classification

Reading Material

Neural Network Basics (Jan 15)

Content

  • Cross Entropy Loss
  • Gradient Descent
  • Components of a feedforward neural network

Slides

Neural Network Basics

Reading Material

Word Vectors (Jan 17)

Content

  • Deep Averaging Network for Text CLassification
  • Lexical Semantics
  • Distributional Semantics
  • Evaluating Word Vectors

Slides

Word Vectors

Reading Material

Language Modeling (Jan 22)

Content

  • What is a language model
  • How to evaluate a language model
  • How to build a language model - N-gram language model, a simple feedforward neural LM

Slides

Language Modeling

Reading Material

[Eisenstein 6.1-6.2, 6.4]

Language Modeling (Jan 24)

Content

  • Feedforward Language Model
  • Recurrent Neural LM, Attention
  • Building blocks of a transformer

Slides

Neural LM

Reading Material

[J&M Chapter 8, 9]

[Eisenstein 6.3]

[Luong15]

[Illustrated Transformer]

Transformers (Jan 29)

Content

  • Self attention
  • Transformer Encoder
  • Transformer Decoder (Cross Attention, Masked Self Attention)
  • Impact of transformers

Slides

Transformers

Reading Material

[Illustrated Transformer]

[J&M Chapter 9]

[Attention is all you need]

Tokenization (Jan 31)

Tokenization Contd. / Masked LMs (February 7)

Content

  • Unigram tokenizer
  • Pretraining / finetuning paradigm
  • Masked LMs - BERT, RoBERTa, ELECTRA

Slides

Masked LMs

Reading Material

[Illustrated BERT)]

[BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding]

[RoBERTa]

[ELECTRA]

Pretraining II (February 12)

Content

  • T5 / BART / UL2 / GPT2
  • Decoding strategies

Slides

Pretraining II

Reading Material

[What happend to BERT/T5]

[Decoding strategies]

Pretraining III (February 14)

Content

  • Scaling
  • Prompting
  • In-context learning
  • CoT

Slides

Pretraining II

Reading Material

Instruction Following (February 19)

Content

  • Instruction Tuning (T0, FLAN)
  • Evaluating Instruction Tuned LMs
  • Basics of RLHF

Slides

Instruction Following

Reading Material

Preference Optimization (February 21)

Content

  • Reward Modeling
  • Basics of RLHF
  • Direct Preference Optimization

Slides

Learning from Preferences

Reading Material

[Illustrating Reinforcement Learning from Human Feedback (RLHF)]

[DPO]

[PPO vs DPO]

[Other resources]

Parameter Efficient Finetuning (February 26)

Content

  • LoRA
  • QLoRA

Slides

(q)lora

Reading Material

[LoRA]

[QLoRA]

Evaluation (February 28)

Content

  • What is Benchmarking
  • Open and close ended evaluation
  • LLM Evaluation Challenges

Slides

benchmarking

Reading Material

[The Evolving Landscape of LLM Evaluation (for Quiz)]

Sequence Tagging (March 5)

Content

  • Why sequence tagging
  • HMMs
  • Viterbi

Slides

sequence tagging

Reading Material

TBA

Parsing (March 7)

Content

  • Constituency Parsing
  • CKY Algorithm
  • Dependency Parsing (Intro)
  • Semantic Parsing (Into)

Slides

Parsing

Reading Material

TBA

Spring Break (March 12 & 14)

No Class

Interpretability (March 19)

Content

  • Global vs Local Explanation
  • Post hoc explanations (LIME, Gradient-based)
  • Probing

Slides

Interpret

Reading Material

TBA

Efficiency (March 21)

Content

  • Speculative Decoding, Flash Attention
  • Quantization, Pruning, Distillation

Slides

Efficiency

Reading Material

TBA

Multimodality (March 26)

Content

  • ViT
  • CLIP
  • Image + Text -> Text

Slides

Multimodal

Reading Material

Multilinguality (March 28)

Content

TBD

Slides

Multilingual

Reading Material

TBD