Safe and Scalable Web Agent Learning via Recreated Websites

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chae, Hyungjoo, Park, Jungsoo, Ritter, Alan
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912961166049280
author Chae, Hyungjoo
Park, Jungsoo
Ritter, Alan
author_facet Chae, Hyungjoo
Park, Jungsoo
Ritter, Alan
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2603_10505
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Safe and Scalable Web Agent Learning via Recreated Websites
Chae, Hyungjoo
Park, Jungsoo
Ritter, Alan
Computation and Language
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.
title Safe and Scalable Web Agent Learning via Recreated Websites
topic Computation and Language
url https://arxiv.org/abs/2603.10505