Description

The “large” in LLM is more foundational than descriptive: models improve predictably as they grow. Increases in parameters and data lead to reliable increases in accuracy. Recent results from OpenAI demonstrate that a third axis, test-time compute, also exhibits similar properties in specific domains. While the details of this method are not yet known, the result is critical for future LLM design. This survey talk will introduce the literature related to test-time compute and model self-improvement, and will discuss the expected implications of test-time scaling. This talk will also briefly connect these research directions to current open-source efforts to build effective reasoning models.

Alexander “Sasha” Rush is a professor at Cornell Tech and a researcher at Hugging Face. He received his PhD from MIT in 2014. He was a postdoc at Facebook Artificial Intelligence Research (FAIR) with Yann LeCun and an assistant professor at Harvard University. His research interest is in the study of language models with applications in novel architectures, efficient inference, and scaling. In addition to research, he has written several popular open-source software projects supporting LLM research and programming for deep learning. He is a founder of COLM and a former secretary for ICLR. His projects have received paper and demo awards at major NLP, ML, visualization, and hardware conferences, an NSF CAREER Award, and a Sloan Research Fellowship.


 

The Richard M. Karp Distinguished Lectures were created in Fall 2019 to celebrate the role of Simons Institute Founding Director Dick Karp in establishing the field of theoretical computer science, formulating its central problems, and contributing stunning results in the areas of computational complexity and algorithms. Formerly known as the Simons Institute Open Lectures, the series features visionary leaders in the field of theoretical computer science and is geared toward a broad scientific audience.

Light refreshments will be available at 10:30 a.m., prior to the start of the lecture. 

The lecture recording URL will be emailed to registered participants. This URL can be used for immediate access to the livestream and recorded lecture. Lecture recordings will be publicly available on SimonsTV about 12 to 15 days following each presentation unless otherwise noted.

The Simons Institute regularly captures photos and video of activity around the Institute for use in publications and promotional materials. 

If you require special accommodation, please contact our access coordinator at simonsevents@berkeley.edu with as much advance notice as possible.

All scheduled dates:

Past

No Past activities yet

Register

Registration is required to attend this event in person, for access to the livestream, or for early access to the recording. Seating is first come, first served.

If you require special accommodation, please contact our access coordinator at simonsevents@berkeley.edu with as much advance notice as possible.

Register Now