Theory and Practice of Efficient Evaluation

Speaker: Mr. Vilém ZOUHAR
ETH Zürich

Title: Theory and Practice of Efficient Evaluation

Date: Monday, 2 November 2026

Time: 11:00am - 12:00noon

Venue: Lecture Theater B (Lam Woo Lecture Theater), HKUST

Abstract:

Science without evaluation is limited to blind exploration, though good evaluation comes at a cost. As AI reaches human parity, previous evaluation practices are becoming obsolete. We cover various perspectives on AI evaluation, ranging from human to automatic evaluation and from theoretical to practical aspects, such as how to construct challenging and efficient testsets.

Biography:

Vilém is last-year PhD at ETH Zurich and a Google PhD Fellow. He works on both the practical and theoretical aspects of AI evaluation (human annotations, automated metrics, methods), especially in the context of multilingual NLP. He organizes WMT human evaluation and recently the Last Translation Benchmark efforts.