1. events
  2. all
  3. 2026

ECE Seminar Series – Feb 20 (Fri) @2:00pm: “Building AI Agents that Reason, Act, and Evolve,” Xin (Eric) Wang, Assistant Professor, Computer Science, UCSB

Date and Time
photo of eric wang

Location: Henley Hall (HH), Room 1010
Come at 1:30p for Cookies, Coffee and Conversation
DISTINGUISHED LECTURE at the ECE SEMINAR SERIES

Abstract

In this talk, I will present our recent progress toward building general-purpose AI agents that can reason, act, and evolve in complex real-world environments. I will begin with the challenge of multimodal reasoning: enabling models to carry out fluid, abstract deliberation while remaining grounded in visual evidence, so intermediate thinking is both flexible and verifiable.

I will then show how we translate reasoning into reliable action by developing computer-use agents that can autonomously operate graphical user interfaces (GUIs). These agents tackle long-horizon tasks through improved memory and planning, stronger grounding to dynamic interfaces, and scalable evaluation and selection strategies, achieving human-level computer use across everyday and enterprise workflows.

Finally, I will briefly highlight our direction on open-ended self-improvement with group-evolving agents, where experience sharing within a group—treated as the evolutionary unit—turns exploration into sustained progress, pointing toward agents that can continually improve beyond fixed, human-designed agentic frameworks.

Bio

Xin (Eric) Wang is an Assistant Professor in the Computer Science Department at UC Santa Barbara and the Head of Research at Simular. He also directs the UCSB Center for Responsible Machine Learning (CRML). His research interests include Natural Language Processing, Computer Vision, and Machine Learning, with an emphasis on Multimodal and Agentic AI. Previously, he was a faculty at UC Santa Cruz and worked at Google Research, Meta FAIR, Microsoft Research, and Adobe Research.

Eric has served as (Senior) Area Chair for conferences such as ACL, NAACL, EMNLP, ICLR, NeurIPS, and ICML, and organized workshops and tutorials at those venues. He has received several awards and recognitions for his work, including Best Paper Awards from CVPR and ICLRW, Google Research Faculty Award, Amazon Alexa Prize Awards, JPMorganChase Faculty Research Award, Cisco Faculty Research Awards, eBay Faculty Research Awards, AAII Interdisciplinary Research Award, and various gift awards/grants from Adobe, Apple, Amazon, Snap, Microsoft, OpenAI, Cybever, Orby, etc.

Hosted by: Lecture at the ECE Seminar Series

Submitted by: Prof. Haewon Jeong <haewon@ece.ucsb.edu>