Datasets:
language:
- en
license: cc-by-4.0
size_categories:
- 1K<n<10K
task_categories:
- text-generation
pretty_name: UnCoBench
tags:
- text
- benchmark
- multi-agent
- collaboration
- reasoning
- synthetic
- evaluation
UnCoBench
UnCoBench evaluates how well a main agent completes collaborative tasks when other agents exhibit unreliable behavior. The dataset contains 1,000 groups of paired English tasks across five settings, yielding 5,000 task instances.
The main agent must gather information from role agents, resolve gaps or inconsistencies, and produce a final decision and, when required, a numerical result. Role agents have separate descriptions and private information. Shared task IDs make it possible to compare performance across collaboration settings.
Dataset Overview
| Property | Description |
|---|---|
| Language | English |
| Task groups | 1,000, identified by task_001 through task_1000 |
| Task instances | 5,000: 1,000 in each setting |
| Interaction | A main agent queries role agents and submits a final answer |
Collaboration Settings
The four primary settings support paired comparisons of individual unreliable behaviors. The additional mixed setting combines them as a stress test.
| Setting | Description |
|---|---|
reliable |
Baseline collaboration without the three injected unreliable behaviors. |
misaligned |
A required record in the public candidate directory points to a different context. The appropriate record remains available, requiring the main agent to recognize and resolve the mismatch. |
incomplete |
A necessary numerical answer is initially omitted or given as a truthful but insufficient bound or interval. A follow-up can recover the precise value. |
missed |
A required role initially does not respond. The main agent must follow up in a later round to obtain an answer. |
mixed |
Combines the three unreliable behaviors across three different required roles. |
License
The benchmark data is licensed under CC BY 4.0, consistent with the data license. When redistributing the data, credit the UnCoBench contributors, include the license, and indicate any modifications.