Abstract: As students increasingly turn to Large Language Models (LLMs) for instant answers and fast coding, we are witnessing a measurable decline in crucial study…
Abstract: How do we find a high-value matching when we are unsure which matches are viable? This challenge is central to many online platforms, from…
Abstract: Diffusion models have rapidly advanced to become a central generative engine in modern AI. Yet two major challenges remain: enhancing test-time compute scalability and…
Abstract: As AI agents take on increasingly complex tasks, from business process automation to cloud infrastructure management, their effectiveness hinges on the ability to learn…
UC Irvine Informatics and Computer Science Distinguished Talk
Abstract: In this talk, I will share our work on designing “everyday” accessible experiences that bridge physical gaps and overcome perceptual barriers.
Abstract: Local search is a powerful heuristic for solving hard optimization problems, used for tasks ranging from training neural networks to finding Nash equilibria. In…
Abstract: Today’s cloud and machine learning workloads are placing ever-increasing demands on datacenter operators in terms of scale and link rates. Optical circuit switches have…
Abstract: There has been a lot of exciting recent progress on new and powerful machine learning algorithms and architectures: how to learn. But for autonomous…
Abstract: Quantum computing is error-prone and needs error correction. Near-term quantum computing of today does not have enough qubits for incorporating quantum error correction. In…
Abstract: Complex organisms function through many types of interactions. Thanks to advancements of experimental technologies and accumulation of resulting data, it is now possible to…
Abstract: A natural intuition in AI is that smart agents should tackle hard problems by building on solutions to easier ones. This idea has inspired…
Abstract: In a digraph, a dicut is a cut where all the arcs cross in one direction. A dijoin is a subset of arcs that…