NUWA 风险评测 · 评测基准
PortBench
NUWA 研究解读 · 独立复核待完成
NUWA Research Note
研究问题研究设计与假设来源发现NUWA 解读局限与相反证据研究与决策意义机构与角色原始资源模型辅助整理 · 选读原始全文或官方文章中的相关章节；尚未完成逐章节审计。
研究速读
NUWA 解读：百分比汇总分级移植表现，不一定是完整解题比例。
研究问题
智能体能否依据参考实现跨语言移植命令行工具？
研究设计与假设
CAISI私有基准测CLI跨语言移植，使用Inspect ReAct与百万加权令牌预算。
来源发现
报告列GPT-5.5为78%，Opus 4.6为60%，DeepSeek V4 Pro为44%。
PortBench连续计分，未纳入二元成功成本比较。
NUWA 解读
NUWA 解读：百分比汇总分级移植表现，不一定是完整解题比例。
局限与相反证据
官网表示将另行详述，现来源未披露题数、评分粒度与完整验证。
研究与决策意义
模型建议：明确标注连续分，不作二元成功或成本等价解释。
精确来源位置
- PortBench · Capability Results: Figure 2Selected original resource https://www.nist.gov/news-events/news/2026/05/caisi-evaluation-deepseek-v4-pro; sha256:37f75aab5640e04d198b94a0ca57704b06d7823caeeb05c4280c8fc461b00623; archived source capture; exact capture time not independently established
- PortBench · Benchmarks: PortBenchSelected original resource https://www.nist.gov/news-events/news/2026/05/caisi-evaluation-deepseek-v4-pro; sha256:37f75aab5640e04d198b94a0ca57704b06d7823caeeb05c4280c8fc461b00623; archived source capture; exact capture time not independently established
- PortBench · Capability Lag Measurement: token budgetsSelected original resource https://www.nist.gov/news-events/news/2026/05/caisi-evaluation-deepseek-v4-pro; sha256:37f75aab5640e04d198b94a0ca57704b06d7823caeeb05c4280c8fc461b00623; archived source capture; exact capture time not independently established
- PortBench · DeepSeek V4 Costs Less Than Other Models Of Similar Capability: exclusionsSelected original resource https://www.nist.gov/news-events/news/2026/05/caisi-evaluation-deepseek-v4-pro; sha256:37f75aab5640e04d198b94a0ca57704b06d7823caeeb05c4280c8fc461b00623; archived source capture; exact capture time not independently established
来源核查、数据提取、编辑批准与实验复现分别记录。 方法说明 →
机构与角色
评测协议
引用与复用
引用研究发现时请引用原始出版物；引用 NUWA 的整理与解释时，单独引用本站记录及对应快照。
原始出版物 · BibTeX
@misc{portbench,
title = {PortBench},
url = {https://www.nist.gov/news-events/news/2026/05/caisi-evaluation-deepseek-v4-pro},
year = {2026},
date = {2026-05-01},
version = {2026-10-25 source capture},
note = {2026-10-25 source capture}
}
NUWA 记录 · BibTeX
@misc{portbench-nuwa,
title = {PortBench — NUWA research record},
url = {https://whitzard.tech/nuwa/whitzard-index/work/portbench/},
version = {2026-10-29.1},
organization = {NUWA Frontier AI Safety Lab},
howpublished = {Retained note edition: https://whitzard.tech/nuwa/whitzard-index/data/record-editions/2026-10-29.1--portbench.json},
note = {NUWA research note. Source-reading scope and independent review status are stated on the record. Snapshot 2026-10-29.1. Retained note edition: https://whitzard.tech/nuwa/whitzard-index/data/record-editions/2026-10-29.1--portbench.json.}
}
证据快照 2026-10-29.1 · 索引更新 · RSS
参考文献
- [1] PortBench. 2026-05-01 ; Capability Results: Figure 2; Benchmarks: PortBench; Capability Lag Measurement: token budgets