| |
|
| |
Presented by EvalScience, Agent Evaluation Science NYC 2026 is a one-day symposium bringing together researchers and practitioners working on how AI agents should be evaluated.
Featured speakers and contributors include Hanna Wallach (Microsoft Research), Jason Weston (Meta / NYU), Yoav Artzi (Cornell Tech), Nouha Dziri (Cohere Labs), Peter Henderson (Princeton), Olawale Salaudeen (UIUC), Rachneet Kaur (J.P. Morgan), Ida Momennejad (Microsoft Research), and Saurabh Jha (IBM Research).
We will discuss what agent evaluations should measure, how reliability and validity should be established, where benchmarks and graders fail, how evaluation infrastructure should be built, and what changes when agents are evaluated in interactive and real-world settings.
The program includes invited talks, contributed research presentations, posters, a fireside conversation, and extended discussion and networking.
November 20, 2026 New York City
Registration is free and approval-based. We welcome researchers, builders, and practitioners working on evaluation methods, benchmarks, agent behavior, evaluation infrastructure, and real-world evaluation problems.
We are also accepting contributions including ongoing or published research, evaluation studies, tools, failure analyses, and lessons from evaluating deployed systems.
Learn more: https://evalscience.org/2026fall-nyc/
|
|
|
|
|
|
|
|
