Paper Detail

Safe and Scalable Web Agent Learning via Recreated Websites

Hyungjoo Chae, Jungsoo Park, Alan Ritter

arxiv Score 11.5

Published 2026-03-11 · First seen 2026-03-27

Research Track B · General AI

Abstract

Training autonomous web agents is fundamentally limited by the environments they learn from: real-world websites are unsafe to explore, hard to reset, and rarely provide verifiable feedback. We propose VeriEnv, a framework that treats language models as environment creators, automatically cloning real-world websites into fully executable, verifiable synthetic environments. By exposing controlled internal access via a Python SDK, VeriEnv enables agents to self-generate tasks with deterministic, programmatically verifiable rewards, eliminating reliance on heuristic or LLM-based judges. This design decouples agent learning from unsafe real-world interaction while enabling scalable self-evolution through environment expansion. Through experiments on web agent benchmarks, we show that agents trained with VeriEnv generalize to unseen websites, achieve site-specific mastery through self-evolving training, and benefit from scaling the number of training environments. Code and resources will be released at https://github.com/kyle8581/VeriEnv upon acceptance.

Workflow Status

Review status
pending
Role
unreviewed
Read priority
soon
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{chae2026safe,
  title = {Safe and Scalable Web Agent Learning via Recreated Websites},
  author = {Hyungjoo Chae and Jungsoo Park and Alan Ritter},
  year = {2026},
  abstract = {Training autonomous web agents is fundamentally limited by the environments they learn from: real-world websites are unsafe to explore, hard to reset, and rarely provide verifiable feedback. We propose VeriEnv, a framework that treats language models as environment creators, automatically cloning real-world websites into fully executable, verifiable synthetic environments. By exposing controlled internal access via a Python SDK, VeriEnv enables agents to self-generate tasks with deterministic, p},
  url = {https://arxiv.org/abs/2603.10505},
  keywords = {cs.CL},
  eprint = {2603.10505},
  archiveprefix = {arXiv},
}

Metadata

{}