

Evaluating LLM Systems
Hosted by WiDS Israel
Registration
Past Event
About Event
Evaluating LLM Systems
Measuring Quality without Perfect Ground Truth: Techniques for assessment using LLM-as-a-judge, pairwise comparisons, and automated evaluation frameworks.
Failure Mode Analysis & Hallucination Mitigation: Identifying common pitfalls and implementing strategies to reduce factual errors.
Operational Best Practices: Environment separation, robust logging, and advanced tracing utilizing LangSmith.
Join Zoom Meeting
https://lsports.zoom.us/j/84834024082?pwd=P8baI1ka97lC4h73iTb7ujpiyzjMn8.1
Location
https://lsports.zoom.us/j/84834024082?pwd=P8baI1ka97lC4h73iTb7ujpiyzjMn8.1