The UC Santa Cruz Natural Language Processing (NLP) Master’s program is proud to highlight the significant role our faculty, staff, and students played in the recent BayLearn 2025 symposium, a premier annual event that brings together machine learning researchers and practitioners from across the Bay Area’s top academic institutions and industries. Many of the leading Machine Learning researchers in the Bay Area spoke at this event, including Chris Manning, Bryan Catanzaro, Ronan Collobert, Mingqiu Wang, and Ed Chi.
Faculty and Graduate Students Leading the Charge
The NLP MS program’s strength lies in the groundbreaking research conducted by our faculty and their labs. Two UCSC PhD students, who serve as TA’s and GSI for the NLP program, were selected to present their latest findings, sharing their work with leading minds in the field.
Faculty Organizers
UCSC took a major role in the planning and production of this year’s symposium. Prof. Ian Lane and Prof. Leilani Gilpin served on the BayLearn 2025 organizing committee, helping to shape the agenda, secure speakers, and manage the overall success of the event. Lane and his staff, Kristina Alvarez and Lisa Schaffer, were highly involved in the planning and production process as Host Leaders of the event . Their involvement underscores the leadership and influence UCSC faculty hold within the broader Bay Area machine learning community.
Graduate Student Presenters
Sicong Huang, a CSE PhD student and perennial NLP MS TA, presented his paper, “Context, Models and Prompt Optimization for Automated Hallucination Detection in LLP Output” The findings in this award winning paper, co-authored by Jincheng He, Shiyuan Huang, Karthik Raja Anandan, Arkajyoti Chakraborty and Ian Lane, are vital in building more trustworthy Generative AI. This paper won the Best System Description Award Task 3 at SemEval 2025, and was the basis for the team’s submission in the “Multilingual Shared-task on Hallucinations and Related Observable Overgeneration Mistakes” task at SemEval 2025, where the team ranked first in average IoU across all languages and ranked in the top two systems for 11 of the 14 evaluation languages.
Diji Yang, a CSE PhD and NLP MS GSI, presented his paper, “Knowing You Don’t Know: Learning When to Continue Search in Multi-round RAG through Self-Practicing.” This paper, co-authored with Linda Zeng, Jinmeng Rao and Yi Zhang, teaches RAG systems to be smarter about when to search for answers.
These presentations showcased the innovative research coming out of UCSC, covering topics at the intersection of language, AI, and systems, and generating valuable discussion among attendees. The opportunity to present at BayLearn gives our students critical exposure and networking opportunities with peers and industry leaders.
Why This Matters for UCSC NLP
The strong presence at BayLearn 2025 demonstrates that UCSC’s NLP program is not just keeping pace with the industry—we’re helping to set its direction. For current and prospective students, this high-level engagement means:
- Access to Top Faculty: Learn directly from professors like Ian Lane and Leilani Gilpin, who are actively shaping the biggest regional machine learning conferences.
- Exposure and Networking: Opportunities to present your work and connect with researchers from Google, Meta, Stanford, and other Bay Area powerhouses.
We are excited to continue pushing the boundaries of NLP and machine learning and look forward to an even stronger presence at BayLearn 2026!