scale-ai/hil-bench
Coding
Behavior
HiL-Bench (Human-in-the-Loop): tests if agents know when to ask for help rather than proceed with uncertain knowledge.
Run this task
CLI:
inspect eval inspect_harbor/scale_ai_hil_bench --model openai/gpt-5Python:
from inspect_ai import eval
from inspect_harbor import scale_ai_hil_bench
eval(scale_ai_hil_bench(), model="openai/gpt-5")Dataset information
| Harbor registry | scale-ai/hil-bench |
| Inspect task | scale_ai_hil_bench |
| Latest digest | sha256:a308c71edf51736003412b8353cfb25f0cdfd58065535e18e2e8937fe6f7ac42 |
| Samples | 600 |
| Paper | arxiv |
| Source | https://labs.scale.com/papers/hil |
See Task Parameters for the parameter set shared across all Harbor tasks.