Link Search Menu Expand Document
CSCI 2952-N

Course Schedule

Paper reading list and presenters

Jan 28, Tue
Course Overview (Slides)
Chen
  1. How to Read a CS Research Paper by Philip Fong
  2. How to do research by Bill Freeman
  3. How to do write a good paper by Bill Freeman
  4. Novelty in Science by Michael Black
  5. How to speak (video) by Patrick Winston
Jan 30, Thu
Deep Learning Recap (Slides)
Chen
Feb. 3, Mon
Due Presentation signup sheet
Feb. 4, Tue
Learning with Various (or “No”) Supervision (Slides)
Chen
Feb. 6, Thu
Self-supervised Learning and its Emerging Abilities (Reading survey / Slides / Questions)
Alex, Bumjin, Nuo Wen, Preetish
  1. Emergent Abilities of Large Language Models
  2. Are Emergent Abilities of Large Language Models a Mirage?
Feb. 11, Tue
Few-shot and In-context Learning (Reading survey / Slides / Questions)
Evani, John, Nathan, Taishi, Winston
  1. Matching Networks for One Shot Learning
  2. Language Models are Few-Shot Learners
Feb. 13, Thu
Transformer and its variants (Reading survey / Slides / Questions)
Hayden, Nicholas, Oliver, Yik Siu
  1. Perceiver: General Perception with Iterative Attention
  2. Efficiently Modeling Long Sequences with Structured State Spaces
Feb. 20, Thu
Flow Models (Slides)
Calvin
  1. An Introduction to Flow Matching
  2. Diffusion Meets Flow Matching: Two Sides of the Same Coin
Feb. 20, Thu
MP Mini Project
Due on March 18
  1. Mini Project Handout
  2. Submission Form
  3. Diffusion Notebook
  4. Normalizing Flows Notebook
  5. Flow Matching Notebook
Feb. 25, Tue
Controlling Generative Models (Reading survey / Slides)
Jiayi, Shengmai, Shishi, Yiyang, Zhuoxuan
  1. Adding Conditional Control to Text-to-Image Diffusion Models
  2. SDEdit: Guided Image Synthesis and Editing with Stochastic Differential Equations
Feb. 25, Tue
FINAL Final Project Proposal
Due on March 6
  1. Example Final Projects from Spring 2023
Feb. 27, Thu
Video Generation Meets the Laws of Physics (Reading survey / Slides)
Mindy, Heon, Gaurav, John, Tanish
  1. PhysDiff: Physics-Guided Human Motion Diffusion Model
  2. PhysGen: Rigid-Body Physics-Grounded Image-to-Video Generation
Mar. 4, Tue
Representation Learning, Revisited (Reading survey / Slides)
Jiaqi, Ruiqi, Zihan, Kyle
  1. Moving Off-the-Grid: Scene-Grounded Video Representations
  2. Learning and Leveraging World Models in Visual Representation Learning
Mar. 6, Thu
Recent Advances on Multimodal LLMs (Slides)
Zitian, Shijie
Mar. 11, Tue
Understanding vs. Generation (Reading survey / Slides)
Yiwen, Chen-En, Zhuoyang, Xilin
  1. Autoregressive Image Generation without Vector Quantization
  2. Transfusion: Predict the Next Token and Diffuse Images with One Multi-Modal Model
Mar. 12, Wed
FINAL One-slide Final Project Proposal (Slides)
Due on March 12
Mar. 13, Thu
Final Project Proposal (Slides)
All Students
Mar. 18, Tue
INVITED Guest Lecture by Kaiming He
Mar. 21, Fri
INVITED Invited Talk by Jiatao Gu
Apr. 1, Tue
Recent Advances on Reinforcement Learning (Slides)
Calvin, Zilai
Apr. 3, Thu
World Models (Reading survey / Slides)
Noah, Praccho, Adam, Nicholas
  1. Genie: Generative Interactive Environments
  2. Mastering Diverse Domains through World Models
Apr. 8, Tue
Robot Learning with Video Generative Models (Reading survey / Slides)
Sami, Naicheng, Shangyang, Roger, Wanjia
  1. Learning Universal Policies via Text-Guided Video Generation
  2. Learning Interactive Real-World Simulators
Apr. 11, Fri
INVITED Invited Talk by Jun-Yan Zhu
Apr. 15, Tue
RL for LLMs (Reading survey / Slides)
Dan, Daniel, Jacob, Shane
  1. Direct Preference Optimization: Your Language Model is Secretly a Reward Model
  2. DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
Apr. 17, Thu
LLMs for Reasoning (Slides)
Apoorv
Apr. 22, Tue
Agents (Reading survey / Slides)
Duo, Nikunj, Yixiang
  1. Toolformer: Language Models Can Teach Themselves to Use Tools
  2. Generative Agents: Interactive Simulacra of Human Behavior
Apr. 24, Tue
From GPT to DeepSeek (Reading survey / Slides)
Yanshu, Ziqi, Bozheng, Hongwei, Zhenke
  1. GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding
  2. Beyond Human Data: Scaling Self-Training for Problem-Solving with Language Models
May 9, Fri
Final project presentations (Lubrano 1 to 4 pm) (Slides)
May 9, Fri
Due Project submission (Form)