Paper Detail

SCAFFOLD: Self-Improving Web Agents via Recursive Parametric Skill Abstraction

Bowei He, Xiaokun Zhang, Meng Ding, Xue Liu

arxiv Score 19.5

Published 2026-08-31 · First seen 2026-09-09

Research Track B · General AI

Abstract

Web agents need to navigate visually rich, long-horizon interfaces that change across sites, yet most previous agents still learn each task in isolation and discard the procedural knowledge they accumulate. Recent skill-augmented frameworks take an important first step, but they treat the skill library as a flat or two-tier prompt-side cache and offer no principled mechanism for compressing redundancy or composing skills recursively. We introduce \textsc{Scaffold}, a self-improving framework for visual web agents that (i) induces parametric, executable skills from successful trajectories under a multi-instance abstraction constraint, (ii) maintains a recursively composed hierarchy in which higher-level skills invoke lower-level ones, (iii) compacts the library via a minimum-description-length (MDL) criterion and behavioral equivalence checking, and (iv) periodically distills skill-augmented trajectories back into model weights to internalize the abstractions. Across WebArena, VisualWebArena, and a held-out split of Online-Mind2Web, \textsc{Scaffold} improves success rate by $11.1$--$17.2$ absolute points over the strongest skill-augmented baseline and shows monotonic gains across five self-improvement iterations without library collapse. We release the code and documents in the Github \href{https://github.com/BokwaiHo/SCAFFOLD}{repository}.

Workflow Status

Review status
pending
Role
unreviewed
Read priority
now
Vote
Not set.
Saved
no
Collections
Not filed yet.
Next action
Not filled yet.

Reading Brief

No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.

Why It Surfaced

No ranking explanation is available yet.

Tags

No tags.

BibTeX

@article{he2026scaffold,
  title = {SCAFFOLD: Self-Improving Web Agents via Recursive Parametric Skill Abstraction},
  author = {Bowei He and Xiaokun Zhang and Meng Ding and Xue Liu},
  year = {2026},
  abstract = {Web agents need to navigate visually rich, long-horizon interfaces that change across sites, yet most previous agents still learn each task in isolation and discard the procedural knowledge they accumulate. Recent skill-augmented frameworks take an important first step, but they treat the skill library as a flat or two-tier prompt-side cache and offer no principled mechanism for compressing redundancy or composing skills recursively. We introduce \textbackslash{}textsc\{Scaffold\}, a self-improving framework for},
  url = {https://arxiv.org/abs/2609.05511},
  keywords = {cs.AI, cs.CL},
  eprint = {2609.05511},
  archiveprefix = {arXiv},
}

Metadata

{}