© 2026 Buobe

Digi-Q: Learning VLM Q-Value Functions for Training Device-Control Agents | Buobe

Digi-Q: Learning VLM Q-Value Functions for Training Device-Control Agents

Published by

yifeizhou02.github.io

on •1 min read

While most paradigms for building foundation model agents rely on prompting or fine-tuning on demonstrations, it is not sufficient in dynamic environments (e.g., mobile device control).

Digi Learning Value Functions Training