Workflow-Guided Exploration research code

Research code that learns browser-control policies by constraining exploration with workflows induced from demonstrations, with a Selenium-based MiniWoB environment and experiment harness.

AI-assisted editorial review against a primary source. This is a catalog annotation, not an independent reproduction or endorsement.

What is this resource?

Resource type

Frameworks

Environment

web

Topics

reinforcement-learning, web-navigation, workflow, history

Record publication date

Not recorded

Evidence retrieved

2026-10-17T11:18:16Z

Resource ID

github:stanfordnlp/wge

What licenses are recorded?

Code, training data, weights, and documents can have different terms. A blank declaration means unknown, not unrestricted use.

Code

Apache-2.0 (Stanford WGE source license; not a blanket license for all third-party tasks, embeddings or demonstration resources.)

Data

Not recorded

Document

Not recorded

Weights

Not recorded

License review notes

• Code: Stanford WGE source license; not a blanket license for all third-party tasks, embeddings or demonstration resources. Source 1

Recorded specifications

availability

Public historical training and environment-wrapper source; legacy dependencies documented.

repository revision

bed526657a62a64a163f80deb4a8627cb63871a3

Explore related topics

Scope and limitations

The released repository includes training policies, experiment configuration and a MiniWoB sandbox extended for the 2018 paper. This is the WGE implementation, not another copy of the already cataloged Farama MiniWoB++ benchmark and not the original OpenAI MiniWoB release. Its documented dependency is PyTorch 0.1.12, and current runtime compatibility was not tested. Stanford code declares Apache-2.0; bundled third-party browser tasks and externally downloaded demonstration/embedding data retain their own terms.

Related resources

• documented by: Reinforcement Learning on Web Interfaces Using Workflow-Guided Exploration

Primary evidence and verification

Source retrieval is separate from experiment reproduction. Use the original publication or publisher page when citing claims about the resource.

1. editorial primary review

   Retrieved 2026-10-17T11:18:16Z. Supports: name, kind, summary, environments, topics, scope, notes, relations.

   Content fingerprint

   50ee5234bf62320d096fbe4c11c7be21f50c9e09dae97754a16a5bbde2f1ab19
2. editorial primary review

   Retrieved 2026-10-17T11:18:15Z. Supports: identity.

   Content fingerprint

   66fcd6577d1867abf84e62fe9edf2e3f223292724cbe6b9d0ef02b582db5c8b2
3. editorial primary review

   Retrieved 2026-10-17T11:19:50Z. Supports: release_status, details.repository_revision.

   Content fingerprint

   926aff11e14c529d05d9bc51fe12aadb007818119da55f99c2fa47d0387f466d
4. editorial primary review

   Retrieved 2026-10-17T11:20:17Z. Supports: details, notes.

   Content fingerprint

   249b23ff62e21bd446b3e411b1d7ea7f173c7e19e42ed75e97800ed32201bb3a
5. editorial primary review

   Retrieved 2026-10-17T11:23:05Z. Supports: licenses.code.

   Content fingerprint

   3df3dd1e3149cd6cc1a7b052a237a6653449a7c5f18edf19a3a79294c7d541c1

Suggest a correction

Open a source-backed correction. Include this record ID and the original source. Read our editorial policy and coverage boundaries.