From Prompt To Prediction: Benchmarking & Evaluating Gen AI

How can we measure the quality and reliability of generative models?
Join us at LSEG Bucharest for an evening dedicated to exploring that question.

We’ll kick off with Sanja Huković, Ph.D., MBA, Head of Model Risk at LSEG, who will introduce model risk management principles in AI.
Then, our keynote speaker, Prof. Radu Tudor Ionescu from the University of Bucharest will discuss his latest CVPR 2025 paper, “PQPP: A Joint Benchmark for Text-to-Image Prompt and Query Performance Prediction” ---> the first large-scale benchmark connecting prompt difficulty, query prediction, and human relevance in multimodal AI.

📍 LSEG Bucharest, Bulevardul Iuliu Maniu 6F, București
🗓 14 Nov, Friday 18:15 - 20:00
🕕 Join us from 18:15 for networking & snacks | Kickoff at 18:30