WhitzardNUWA Lab ↗
NUWANUWA INDEX

Search NUWA Index

Research · Benchmarks · Findings · Perspectives

中文
← NUWA Index

Evaluation dossier

Agent-SafetyBench

Evaluates agent safety through interactive environments and tool use, separating agent-level failures from chatbot-only tests.

Tsinghua University / arXiv:2412.14470v2 / Behavior

Sources and provenance ↓ Read the research narrative →

Task and measurement

Benchmark protocol profile

Task / subject
Model or agent system under the original task configuration
Measurement target
Evaluates agent safety through interactive environments and tool use, separating agent-level failures from chatbot-only tests.
Interpretation boundary
Broad safety coverage does not automatically establish a direct link to a specific frontier-risk mechanism.
Measurement dossier · NUWA reading

Source-method reading and retained catalogue information are distinguished. Publication does not establish independent evidence or bilingual editorial review.

Measurement construct
Which safety failures arise when agents interact with environments and tools?
Task population
349 environments and 2,000 cases cover eight risk categories and ten failure modes; existing cases are refined, deduplicated and standardized.
Scoring and failure modes
Case-level safety scores summarized by total/failure mode; helpfulness assessed separately.
System conditions
349 environments, 2,000 cases, sixteen tested agents.
Statistical meaning
Source-reported result statements; this does not establish trial counts, denominators or an estimator: None of sixteen tested agents exceeded 60% total safety score. Defense prompts gave limited benefits and some weaker models did not improve (§4.5).
Validity and interpretation
Constructed scenarios and a particular scorer do not establish real deployment incident rates; higher capability is not a safety guarantee.
Appropriate research use
Model recommendation: inspect failure-mode and helpfulness breakdowns rather than relying on total safety score.
  • Agent-SafetyBench : Evaluating the Safety of LLM Agents · AbstractSelected original resource https://arxiv.org/html/2412.14470; sha256:d75e5fb230a89a9ffa0484fbfe694ac609b7b5de3aa59204cadf73b473737338; archived source capture; exact capture time not independently established
  • Agent-SafetyBench : Evaluating the Safety of LLM Agents · §3.2.2 Refine Existing DatasetsSelected original resource https://arxiv.org/html/2412.14470; sha256:d75e5fb230a89a9ffa0484fbfe694ac609b7b5de3aa59204cadf73b473737338; archived source capture; exact capture time not independently established
  • Agent-SafetyBench : Evaluating the Safety of LLM Agents · §4.2 Main ResultsSelected original resource https://arxiv.org/html/2412.14470; sha256:d75e5fb230a89a9ffa0484fbfe694ac609b7b5de3aa59204cadf73b473737338; archived source capture; exact capture time not independently established
  • Agent-SafetyBench : Evaluating the Safety of LLM Agents · §4.5 Discussion on DefenseSelected original resource https://arxiv.org/html/2412.14470; sha256:d75e5fb230a89a9ffa0484fbfe694ac609b7b5de3aa59204cadf73b473737338; archived source capture; exact capture time not independently established
Original source charts and materials
Agent-SafetyBench — Figure 1
Figure 1 · Original figure 1. Read alongside the protocol scope and limitations below. View original image ↗

Related research questions

References

  1. [1] Zhexin Zhang 1 1 1 Equal contribution., Shiyao Cui 1 1 1 Equal contribution., et al.. Agent-SafetyBench: Evaluating the Safety of LLM Agents. 2024-12-19. v2 [arXiv]; Author affiliations on the original paper title page; Figure 1