Abstract: Likelihood, although useful as a training loss, is a poor search objective for guiding open-ended generation from language models (LMs). Existing generation algorithms must avoid both unlikely strings, which are...
Abstract: Likelihood, although useful as a training loss, is a poor search objective for guiding open-ended generation from language models (LMs). Existing generation algorithms must avoid both unlikely strings, which are...
Title: Scalable and Trustworthy Learning in Heterogeneous Networks
Abstract: To build a responsible data economy and protect data ownership, it is crucial to enable learning models from separate, heterogeneous...
Title: Socially Responsible and Factual Reasoning for Equitable AI Systems
Abstract: Understanding the implications underlying a text is critical to assessing its impact, in particular the social dynamics that...
The Product Management Club meets weekly on Tuesday nights to discuss trends in technology, conduct product thinking seminars, work on case studies, read books, and occasionally prep together for PM recruitment. All UMass...