Expired90% confidenceFactTime unknown
The team conducted a survey of 88 respondents to evaluate AI Agent output quality; aerospace domain experts rated the outputs approximately one grade lower than non-aerospace experts, but still close to the 'good' range.
1
Sources
90%
Confidence
Medium-term (~90 days)
Relevance
6/1/2026
First Seen
Valid until: 8/30/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedIn the experimental design, the AI used an 8:2 split to reserve 20% as an independent test set, and applied 5-fold cross-validation on the 80% training set to avoid data leakage.67% similarUnverifiedMainstream benchmarks for evaluating Agent planning capabilities include ALFWorld, WebArena, and GAIA, and the scores of current state-of-the-art models remain far below human-level performance.66% similarUnverified讲者认为评估AI编程代理应关注80%、90%、99%成功率下的数据而非常被引用的50%成功率版本，因为99%才意味着可信赖并委托给模型64% similarUnverifiedThe average human score on the ARC-AGI test is approximately 85%, while top AI models have long hovered in the 30%–60% range.64% similarUnverified综合智能指数由智能体工作、编程、科学推理、通用能力等 9 项独立评估构成，Sol 与 Faber-5 几乎持平仅差 1 分，但任务完成时间减少约 61%，成本约为原来的一半64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/21611API
curl https://kongchang.com/api/v1/knowledge/claims/21611MCP
get_claim(id=21611)