Gemini 1.5
Google DeepMind — Gemini
rich detail 11 method fields published
Evaluation
Image-to-text refusal evaluation (grounded/ungrounded)
How effectively models refuse sensitive questions about people in images (e.g., religion from a headshot), comparing grounded vs ungrounded queries using MIAP dataset and curated supplements; SxS quality and refusal rates
How it was run
Open a release to see the setup, scoring, and other details that the lab published. Only fields the lab actually disclosed are shown.