policy-violation-benchmark/VIOLA
VIOLA Dataset Description VIOLA is a curated benchmark for evaluating whether LLM-based agents comply with explicit behavioral policies in a multi-agent pipeline. Each example is a single agent execution trace — a real run of the CUGA multi-agent system on AppWorld tasks — in which the target agent's system prompt was modified to induce a specific policy violation using the contrary instruction injection pattern: rather than removing a policy, the distorter… See the full description on the dataset page: https://huggingface.co/datasets/policy-violation-benchmark/VIOLA.
| Author / organization | policy-violation-benchmark |
|---|---|
| License (as declared) | cc-by-4.0 |
| Downloads | 150 (as reported by Hugging Face) |
| Likes | 0 |
| Created | 3 May 2026 |
| Last updated at source | 7 May 2026 |
Popularity, last 3 days
150 now (+8 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.
Recent changes
- 17 Oct 2026 Description: VIOLA Dataset Description VIOLA is a curated benchmark for evaluating whether LLM-based agents comply with explicit behavioral policies in a multi-agent pipeline. Each example is a single agent execution trace — a real run of the CUGA multi → VIOLA Dataset Description VIOLA is a curated benchmark for evaluating whether LLM-based agents comply with explicit behavioral policies in a multi-agent pipeline. Each example is a single agent execution trace — a real run of the CUGA multi
Changes detected by Silytics between refreshes of the source.
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.