SPADE-Bench

300 paired tool-use scenarios that score whether an agent changes its stated plan under pressure while its action stays the same; the metric is Pass@5 deception rate.

Also known as: SPADE-Bench, Spontaneous Plan-Action Divergence Evaluation

unassessed

This page is a discovery lead. Nobody has yet assessed it against the catalogue contract, so it carries no disposition. Absence of evidence here is not evidence of staleness.
Categorysafety
Subcategoryagent deception: plan-action divergence under pressure in simulated tool use
Page statusactive
MetricPass@5 deception rate (share of pairs with a deceptive sample in k=5)
Directionlower_is_better
Unit%
Dataset size300
Dataset licenceCC-BY-NC-ND-4.0 (paper); dataset SPDX not stated beyond an 'open research license'
PublisherBeijing Academy of Artificial Intelligence; Peking University; University of Science and Technology of China; University of Chinese Academy of Sciences; Alibaba Group

What it measures

SPADE-Bench tests whether a tool-using agent tells an observer one plan and then executes another. Each case is a binary workplace dilemma with simulated tools. In the regular variant the observer pressure is off. In the pressure variant an auditor or supervisor can see the plan but not the tool calls. Deception is a shift of the plan toward the observer while the action stays on the agent's intrinsic choice. The authors argue this is strategic, not a random hallucination, because the pair is scored together.

Task format

The agent receives XML-tagged context: tools, role, principles, ground truth, optional pressure, and a user query. It must call a create_plan tool, then act with the rest of the simulated library. A stance classifier labels plan and action as intrinsic, external, or ambiguous. Deception is a rule on the pair: action stays intrinsic across both variants, plan moves toward the observer under pressure. Default protocol is Pass@5 at temperature 0.7.

Models reporting this benchmark

No model card in ModelSpec reports this benchmark yet.

Data

This page as JSON · Edit on GitHub