Guiding Evolution of Artificial Life Using Vision-Language Models
This work addresses the challenge of open-ended evolution in artificial life for researchers, though it is incremental as it builds on prior ASAL methods.
The paper tackled the problem of automating search in artificial life simulations by introducing ASAL++, a method that uses vision-language models to propose new evolutionary targets based on visual history, resulting in EST promoting greater visual novelty and ETT fostering more coherent evolutionary sequences in Lenia tests.
Foundation models (FMs) have recently opened up new frontiers in the field of artificial life (ALife) by providing powerful tools to automate search through ALife simulations. Previous work aligns ALife simulations with natural language target prompts using vision-language models (VLMs). We build on Automated Search for Artificial Life (ASAL) by introducing ASAL++, a method for open-ended-like search guided by multimodal FMs. We use a second FM to propose new evolutionary targets based on a simulation's visual history. This induces an evolutionary trajectory with increasingly complex targets. We explore two strategies: (1) evolving a simulation to match a single new prompt at each iteration (Evolved Supervised Targets: EST) and (2) evolving a simulation to match the entire sequence of generated prompts (Evolved Temporal Targets: ETT). We test our method empirically in the Lenia substrate using Gemma-3 to propose evolutionary targets, and show that EST promotes greater visual novelty, while ETT fosters more coherent and interpretable evolutionary sequences. Our results suggest that ASAL++ points towards new directions for FM-driven ALife discovery with open-ended characteristics.