SCAFFOLD: Self-Improving Web Agents via Recursive Parametric Skill Abstraction
Read the original on arXiv AI →SCAFFOLD is a self‑improving framework for visual web agents that automatically induces parametric, executable skills from successful trajectories and organizes them into a recursively composed hierarchy. It compresses the skill library using a minimum‑description‑length criterion and behavioral equivalence checks, and periodically distills these skills back into model weights to internalize the abstractions. Experiments on WebArena, VisualWebArena, and Online‑Mind2Web show that SCAFFOLD raises success rates by 11.1–17.2 absolute points over the best skill‑augmented baseline and continues to improve across five self‑improvement iterations without collapsing the library.
Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.