Schedule

When What
Week 1 Fundamentals of LLM generation and basic prompting techniques
Tue, Jan 7 Topics: Generation hyperparameters; zero-shot and few-shot prompting; prompting for code. In-class Notebook Out: Programming Capabilities. Create an ACCESS account immediately.
Fri, Jan 10 Topics: High-level overview of Transformers and GPUs. Reading: StarCoder and Llama. Out: Reading 0. In-class: Inference on GPU.
Week 2  
Tue, Jan 14 Topics: Prompt-writing activity; more prompting techniques. In-class Notebook
Fri, Jan 17 Topics: Introduction to Conversational Models. Due: Reading 0. Reading: InstructGPT and UltraFeedback
Week 3  
Tue, Jan 21 Topics: What do models learn? In-class Notebook Due: Programming Capabilities. Out: Math Word Problems
Fri, Jan 24 Topics: Reading discussion; high-speed programming activity. Due: Reading 1.
Week 4  
Tue, Jan 28 Topics: DeepSeek results
Fri, Jan 31 Topics: Introduction to agents In-Class Notebook No reading this week. Reading: SWE-Bench.
Week 5  
Tue, Feb 4 Topics: The Act pattern In-Class Notebook Due: Math Word Problems
Fri, Feb 7 Topics: Information retrieval. Due: Reading 2. In-class Notebook Reading: s1: simple test-time scaling
Week 6  
Tue, Feb 11 Topics: Neural retrieval. In-class Notebook. Recommended Reading: Speech and Language Processing Chapter 14. Out: Thomas the Travel Agent
Fri, Feb 14 Due: Reading 3. Out: Project Proposal
Week 7  
Tue, Feb 18 Topics: Supervised fine-tuning.
Fri, Feb 21 Topics: More on GRPO. Due: Project Proposal
Week 8  
Tue, Feb 25 Class cancelled. Due: Thomas the Travel Agent
Fri, Feb 28 Topics: Speculative decoding and proxy tuning.
Week 9 Spring Break
Tue, Mar 4 No class
Fri, Mar 7 No class
Week 10  
Tue, Mar 11 Topics: High-performance inference. PagedAttention; LLM.int8. Out: Obscure Questions
Fri, Mar 14 Short class: Khoury College PhD Open House
Week 11  
Tue, Mar 18 Topics: Non-Local Control Flow, Containers
Fri, Mar 21 In-class activity: Programming in a language that you do not know.
Week 12  
Tue, Mar 25 Class cancelled.
Fri, Mar 28 Topics: Synthetic data generation. Self-Instruct; Magicoder; MultiPL-T. Due: Obscure Questions
Week 13  
Tue, Apr 1 Topics: Humans solve LLM benchmarks: ARC-AGI 1 and Social IQa.
Fri, Apr 4 Reading: Demystifying CLIP Data and What Matters When Building vision-language models? (Questions)
Week 14  
Tue, Apr 8  
Fri, Apr 11 In-class project presentations (12 min per group)
Week 15  
Tue, Apr 15 Last day of class. In-class project presentations (12 min per group)
   
Tue, Apr 22 Due: [Project]