CS762 Fall 2025

Tentative Schedule (Subject to Change)

Week	Date	Topic	Reading materials	Assignments
1	Thursday, September 4	Course overview and introduction		Sign up for paper presentations and scribes (link to Google sheet)
2	Tuesday, September 9	Evolution of Neural Architecture (lecture)	D2L Book Chapter 7 & 10
2	Thursday, September 11	Evolution of Neural Architecture II	LLM architectures • Attention Is All You Need (deep dive) • BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding • Transformers are RNNs: Fast Autoregressive Transformers with Linear Attention • Linear-Time Sequence Modeling with Selective State Spaces
3	Tuesday, September 16	Evolution of Neural Architecture III	LLM architectures • Attention Is All You Need • BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding • Transformers are RNNs: Fast Autoregressive Transformers with Linear Attention • Linear-Time Sequence Modeling with Selective State Spaces (deep dive)
3	Thursday, September 18	Evolution of Neural Architecture IV	Vision Transformers • An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale • Hierarchical Vision Transformer using Shifted Windows • End-to-End Object Detection with Transformers • HOW DO VISION TRANSFORMERS WORK? (deep dive)
4	Tuesday, September 23	AI Safety and Alignment (lecture)		Team registration (link to Google sheet)
4	Thursday, September 25	AI Safety and Alignment II	Hallucination & Truthfulness • Hallucination is Inevitable: An Innate Limitation of Large Language Models • Semantic Uncertainty: Linguistic Invariances for Uncertainty Estimation in Natural Language Generation • HaloScope: Harnessing Unlabeled LLM Generations for Hallucination Detection • Steer LLM Latents for Hallucination Detection (deep dive)
5	Tuesday, September 30	No class (use the time to discuss the project proposal with team)		Project proposal submission deadline on October 5 Midnight (download proposal latex template here)
5	Thursday, October 2	AI Safety and Alignment III	Alignment • Deep reinforcement learning from human preferences • Fine-Tuning Language Models from Human Preferences • Training Language Models to Follow Instructions with Human Feedback (InstructGPT) • Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback • Understanding the Learning Dynamics of Alignment with Human Feedback • Generalized Preference Optimization: A Unified Approach to Offline Alignment (deep dive)
6	Tuesday, October 7	Meetings with Instructor to Review Project Proposals (optional)
6	Thursday, October 9	AI Safety and Alignment IV	Alignment: Challenges and Limitations • Alignment faking in large language models • Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback • Challenges and Future Directions of Data-Centric AI Alignment
7	Tuesday, October 14	AI Safety and Alignment IV	Other emergent risks • On Targeted Manipulation and Deception when Optimizing LLMs for User Feedback (deep dive) • Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMs
7	Thursday, October 16	Foundation models I (Lecture)
7	Tuesday, October 21	No class
8	Thursday, October 23	Foundation Models II	Large-scale pre-training • Exploring the Limits of Weakly Supervised Pretraining • Improving Language Understanding by Generative Pre-Training • Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer • Learning Transferable Visual Models From Natural Language Supervision • Language Models are Few-Shot Learners (deep dive)
9	Tuesday, October 28	Foundation Models III	Emergent Behaviors • Scaling Laws for Neural Language Models • Emergent Abilities of Large Language Models • Are Emergent Abilities of Large Language Models a Mirage? (deep dive)
9	Thursday, October 30	Foundation Models IV	Reasoning, RL fine-tuning • Chain of Thought Prompting Elicits Reasoning in Large Language Models • Large Language Models are Zero-Shot Reasoners • Tree of Thoughts: Deliberate Problem Solving with Large Language Models • DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning (deep dive) • Visionary-R1: Mitigating Shortcuts in Visual Reasoning with Reinforcement Learning
10	Tuesday, November 4	Foundation Models V	Parameter efficient fine-tuning (PEFT) • LoRA: Low-Rank Adaptation of Large Language Models • Prefix-Tuning: Optimizing Continuous Prompts for Generation • The Power of Scale for Parameter-Efficient Prompt Tuning • Learning to Prompt for Vision-Language Models • Few-Shot Parameter-Efficient Fine-Tuning is Better and Cheaper than In-Context Learning (deep dive)
10	Thursday, November 6	Continual Learning I (Lecture)
11	Tuesday, November 11	Continual Learning II	• Learning without Forgetting • Incremental Classifier and Representation Learning • Overcoming catastrophic forgetting in neural networks • Dark Experience for General Continual Learning: a Strong, Simple Baseline • A Unified and General Framework for Continual Learning (deep dive)
11	Thursday, November 13	Deep Generative Model I (Lecture)	Goodfellow-Bengio-Courville Chapter 20
12	Tuesday, November18	Deep Generative Model II	Foundations • Denoising Diffusion Probabilistic Models (deep dive) • Improved Denoising Diffusion Probabilistic Models • Denoising Diffusion Implicit Models • Consistency Models
12	Thursday, November 20	Deep Generative Model III	Text-to-image and beyond • High-Resolution Image Synthesis with Latent Diffusion Models • Hierarchical Text-Conditional Image Generation with CLIP Latents • Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding • ImageReward: Learning and Evaluating Human Preferences for Text-to-Image Generation
15	Tuesday, December 9	Final project presentation