REFORM Schedule and Past Meetings
Upcoming and past REFORM sessions, papers, and discussion materials.
Meetings are held every Thursday at 5 PM (CoDa W101).
A list of topics, suggested reading, and session plans is available here.
Sign up to be a discussant here. Goal(s) of the discussant group:
- Prepare a 20–30 minute presentation, accessible to a second-year PhD student, focusing on (a) seeding discussion, (b) identifying gaps and connections, and (c) formulating open problems
- We suggest several papers for each week—more than one can cover thoroughly in a week. Pick a small, focused set of papers and read them thoroughly
- Do a single “deep dive” per week about one subject (this can span multiple papers)
Signing up is a great way to (1) force yourself to engage with the content of the paper, (2) get to know your co-discussant(s), and (3) ensure the success of the reading group.
Upcoming & Past Sessions
| Date | Topic | Resources |
|---|---|---|
| 2024-10-16 | Introduction | Slides |
| 2024-10-23 | Scaling Laws 1 (Training Compute-Optimal Language Models) | Paper Slides |
| 2024-10-30 | Scaling Laws 2 (Explaining Neural Scaling Laws) | Paper Slides |
| 2024-11-06 | Data Selection 1 (Perplexity Correlations, Scaling Laws + Data Filtering) | Paper 1 Paper 2 Slides |
| 2024-11-13 | Data Selection 2 (DsDm, LESS) | Paper 1 Paper 2 Tutorial Slides (DsDm) Slides (LESS) |
| 2024-11-20 | Data Selection 3 (Statistical Theory) | Paper Slides |
| 2024-11-20 | Data Selection 3 (Pruning, Prediction) | Paper 1 Paper 2 |
| 2025-01-22 | Post-training 1 (RLHF, AlpacaFarm) | Paper 1 Paper 2 Slides |
| 2025-01-29 | No meeting (ICML Deadline) | |
| 2025-02-05 | Post-training 2 (Direct methods & Offline RL) | Paper 1 Paper 2 Paper 3 Paper 4 Slides 1 Slides 2 |
| 2025-02-12 | Post-training 3 (DeepSeek) | Paper Slides |
| 2025-02-19 | Post-training 4 (Synthetic Data) | Slides |
| 2025-02-26 | Post-training 4 (Synthetic Data & Self-Improvement) | Paper 1 Paper 2 Slides |
| 2025-03-04 | Post-training 5 (Simplicity) | Paper 1 Paper 2 Slides |
| 2025-03-11 | Post-training 5 (In-Context Learning) | Slides |
| (Between-quarter break) | ||
| 2025-04-09 | Reasoning 1 (Introduction) | Slides |
| 2025-04-16 | Reasoning 2 (STaR) | Paper |
| 2025-04-23 | Reasoning 3 (Process rewards) | Slides Paper 1 Paper 2 |
| 2025-04-30 | Reasoning 4 (More Self-improvement) | Paper |
| (Summer break) | ||
| 2025-10-09 | Post-deployment/Safety 1 (CoT Monitoring) | Paper 1 Paper 2 Slides |
| 2025-10-16 | Post-deployment/Safety 2 (Jailbreaking, Elicitation) | Paper 1 Paper 2 Paper 3 Slides |
| 2025-10-23 | Post-deployment/Safety 3 (Hallucinations) | Paper 1 Paper 2 Slides 1 Slides 2 |
| 2025-10-30 | Post-deployment/Safety 3 (Privacy and Memorization) | Slides 1 Slides 2 Paper 1 Paper 2 |
| 2025-11-06 | Post-deployment/Safety 4 (Emergent Misalignment) | Paper |
| 2025-11-13 | Post-deployment/Safety 5 (Out-of-Context Reasoning) | Slides Paper 1 Paper 2 Paper 3 Paper 4 |
| 2026-01-22 | Introduction + Sharpness and Training Dynamics 1 (Edge of Stability) | Slides Paper Extra Reading 1 Extra Reading 2 |
| 2026-02-05 | Sharpness and Training Dynamics 2 (Sharpness-Aware Minimization) | Slides 1 Slides 2 Paper 1 Paper 2 |
| 2026-02-12 | Overfitting and Generalization 1: Double Descent | Slides 1 Slides 2 Paper 1 Paper 2 |
| 2026-02-19 | Overfitting and Generalization 2: Benign Overfitting | Slides 1 Slides 2 Paper 1 Paper 2 |
| 2026-02-26 | Emergent Abilities 1: Grokking | Slides 1 Slides 2 Paper 1 Paper 2 |