Improve dataset card with description and tags
#2
by
nielsr
HF Staff
- opened
README.md
CHANGED
|
@@ -1,3 +1,38 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: mit
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
task_categories:
|
| 4 |
+
- other
|
| 5 |
+
tags:
|
| 6 |
+
- web-browsing
|
| 7 |
+
- agent-benchmark
|
| 8 |
+
- ai-safety
|
| 9 |
+
- reinforcement-learning
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# WebGames Dataset
|
| 13 |
+
|
| 14 |
+
This dataset contains a collection of challenges designed for evaluating general-purpose web-browsing AI agents. The challenges are designed to be easy for humans but difficult for AI agents, and they are fast and simple to run using client-side state and a single-page JavaScript application. Each task provides a unique password upon successful completion, facilitating easy evaluation.
|
| 15 |
+
|
| 16 |
+
[Project Page](https://webgames.convergence.ai) | [Paper](https://hf.co/papers/2502.18356)
|
| 17 |
+
|
| 18 |
+
## Try it now
|
| 19 |
+
|
| 20 |
+
🎮 [webgames.convergence.ai](https://webgames.convergence.ai)
|
| 21 |
+
|
| 22 |
+
## Run locally
|
| 23 |
+
|
| 24 |
+
```bash
|
| 25 |
+
cd webgames
|
| 26 |
+
pnpm install
|
| 27 |
+
pnpm run dev
|
| 28 |
+
```
|
| 29 |
+
|
| 30 |
+
## Download tasks
|
| 31 |
+
|
| 32 |
+
Tasks are available as a dataset on Hugging Face.
|
| 33 |
+
|
| 34 |
+
Alternatively, you can download them from the webgames website:
|
| 35 |
+
|
| 36 |
+
1. Go to [webgames.convergence.ai?showDownloads=true](https://webgames.convergence.ai?showDownloads=true)
|
| 37 |
+
2. Click the download buttons in the top-right corner (csv or jsonl available)
|
| 38 |
+
3. Verify your agent solutions using `solution in messages[-1]` or equivalent, or use the Inspect AI eval scaffolding in the eval folder.
|