A Heart in the Prompt Makes VLMs Write Recursion

A paper by Google DeepMind and Stanford shows that visual symbols in prompt images can drastically steer VLM behavior: adding a red heart made a model write recursive code with 100% success versus 0% without, revealing surprising gaps between model and human visual conventions.

2026-11-04 ~ 2026-11-04 · 2 related posts

• DeepMind-Stanford paper: a red heart in your prompt flips code from 0% to 100% recursive — rohanpaul_ai · 2026-11-04
• A Red Heart Can Steer Code: DeepMind-Stanford Paper Probes Visual Meaning in VLMs — rohanpaul_ai · 2026-11-04