Talks are 45 minutes long, with 15 minutes for discussion. All talks are posted in Pacific Time. Schedule subject to change.

–
Coffee and Check-in
–
Welcome and Introduction
–
Sparks of Artificial General Intelligence
Yin Tat Lee (Microsoft Research)
–
Break
–
Possible Impossibilities and Impossible Possibilities
Yejin Choi (University of Washington)
–
Lunch (on your own)
–
Towards Reliable Use of Large Language Models: Better Detection, Consistency, and Instruction-Tuning
Christopher D. Manning (Stanford University)
–
Break
–
An observation on Generalization
Ilya Sutskever (OpenAI)
–
Panel Discussion (moderated by Alexei Efros)
–
Reception
–
Coffee and Check-in
–
Understanding the Origins and Taxonomy of Neural Scaling Laws
Yasaman Bahri (Google DeepMind)
–
Scaling Data-Constrained Language Models
Sasha Rush (Cornell University & Hugging Face)
–
Break
–
A Theory for Emergence of Complex Skills in Language Models
Sanjeev Arora (Princeton University)
–
Lunch (on your own)
–
Interpretability via Symbolic Distillation
Miles Cranmer (Flatiron Institute)
–
Break
–
Build an Ecosystem, Not a Monolith
Colin Raffel (University of North Carolina & Hugging Face)
–
How to Use Self-Play for Language Models to Improve at Solving Programming Puzzles
Adam Tauman Kalai (Microsoft)
–
Coffee and Check-in
–
Large Language Models Meet Copyright Law
Pamela Samuelson (UC Berkeley)
–
Panel Discussion (moderated by Shafi Goldwasser)
–
Break
–
On Localization in Language Models
Yonatan Belinkov (Technion - Israel Institute of Technology)
–
Lunch (on your own)
–
Language Models as Statisticians, and as Adapted Organisms
Jacob Steinhardt (UC Berkeley)
–
Break
–
Are Aligned Language Models “Adversarially Aligned”?
Nicholas Carlini (Google DeepMind)
–
Formalizing Explanations of Neural Network Behaviors
Paul Christiano (Alignment Research Center)
–
Coffee and Check-in
–
Meaning in the age of large language models
Steven Piantadosi (UC Berkeley)
–
Integrating Language into Intelligent Architectures
Lio Wong (MIT) & Alex Lew (MIT)
–
Break
–
Beyond Language: Scaling up Robot Ontogeny
Jitendra Malik (UC Berkeley)
–
Lunch (on your own)
–
Are LLMs the Beginning or End of NLP?
Dan Klein (UC Berkeley)
–
Break
–
Human-AI Interaction in the Age of Large Language Models
Diyi Yang (Stanford University)
–
Watermarking of Large Language Models
Scott Aaronson (UT Austin & OpenAI)
–
Coffee and Check-in
–
In-Context Learning: A Case Study of Simple Function Classes
Gregory Valiant (Stanford University)
–
Pretraining Task Diversity and the Emergence of Non-Bayesian In-Context Learning for Regression
Surya Ganguli (Stanford University)
–
Break
–
Are Transformers necessary? A data-centric view on generalization
Ludwig Schmidt (University of Washington)
–
Lunch (on your own)
–
Short Talks
Gasper Begus, Bob Carpenter, Fereshte Khani, Song Mei, Weijie Su, Jia Xu
–
Break
–
Short Talks
Nayoung Lee, Shishir Patil, Eric Wallace