WebRewardBench
An offline preference-evaluation set for web process reward models, pairing page and trajectory context with preferred and rejected candidate decisions.
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
- Benchmarks
- Environment
- web
- Topics
- web-navigation, reward-models, evaluation
- Record publication date
- Not recorded
- Evidence retrieved
- 2026-10-17T02:44:50Z
- Resource ID
- url:ac6d0f80765f5a32255a
What licenses are recorded?
Code, training data, weights, and documents can have different terms. A blank declaration means unknown, not unrestricted use.
- Code
- Not recorded
- Data
- Not recorded
- Document
- Not recorded
- Weights
- Not recorded
Evaluation protocol
- Focus
- web process-reward evaluation
- Setting
- offline
- Observations
- page screenshots, text observations, action history, task checklists
- Actions / outputs
- candidate-decision preference
- Evaluation
- Compare reward-model preferences with the provided chosen/rejected decision labels; no independent score was reproduced.
Scale and counting units
- 776 rows — Hub test split source
The current Hub row count and project case-count claim use unresolved counting units. Candidate lists mean a row must not be assumed to equal one task or one pair.
Compare source-reviewed profiles
Recorded specifications
- disabled
- false
- download url
- https://huggingface.co/datasets/LangAGI-Lab/WebRewardBench/tree/main
- gated
- false
- hub id
- LangAGI-Lab/WebRewardBench
Explore related topics
Scope and limitations
Two public Parquet shards are present. Card metadata reports 776 test rows, whereas the project describes 1,000-plus test cases without reconciling that count to rows. Both observations are retained as a grain/coverage caveat rather than treating either as a confirmed number of independent tasks. No data license is declared.
Related resources
- documented by: Web-Shepherd: Advancing PRMs for Reinforcing Web Agents
Primary evidence and verification
Source retrieval is separate from experiment reproduction. Use the original publication or publisher page when citing claims about the resource.
- editorial primary review
Retrieved 2026-10-17T02:44:50Z. Supports: name, kind, summary, environments, topics, scope, relations, details.
Content fingerprint
0968f60eea10b95f79cd16ad3746b0e0b1acaeb0e97c45856f709d53a14dd73a - editorial primary review
Retrieved 2026-10-17T02:44:50Z. Supports: identity, release_status, details, licenses.data.
Content fingerprint
db06c66b8cf5687008b0ad0d9ae269fcc27936887d238cfe2bdeb52151e1683e - editorial primary review
Retrieved 2026-10-17T02:44:04Z. Supports: summary, environments.
Content fingerprint
07431058d8b9a71b80e75b7a74a40a8346c1503eec240628459cc4d4e2f0802f
Suggest a correction
Open a source-backed correction. Include this record ID and the original source. Read our editorial policy and coverage boundaries.