arXiv cs.AI· Hang Xiao, Chuhong Xu, Kainan Zhou, Gangzhen Qian, Lu Yi·· 6 天前AI 评分20
协议选择对基于特征的硬件木马检测的影响：Trust-Hub 家族间的评测边界研究
Protocol effects on feature-based hardware-Trojan detection across Trust-Hub families
AI 导读
研究测量了 Trust-Hub 基准中宿主电路复用对硬件木马检测器表现的影响，而非提出新分类器。在 49,124 个门、16 个网表、5 个宿主家族上，随机森林 F1/AP 从池化测试的 0.914/0.978 降至留出单个网表的 0.636/0.851，再到留出整个宿主家族的 0.460/0.577；XGBoost 从 0.946/0.976 降至 0.464/0.544。
正文 · 原文
Abstract:Trust-Hub reuses host circuits: several files differ mainly in the inserted Trojan. When gates from sibling variants enter both training and test folds, a detector can benefit from host logic it has already seen. We measure that effect instead of proposing another classifier. The corpus contains 49,124 gates from 16 netlists grouped into five host families. We left the parser, 36 gate features, class weighting, model settings, threshold, and family-level aggregation unchanged and altered one choice: the test boundary. The three settings draw test gates from the pooled corpus, withhold a complete netlist, or withhold every variant of one host. The choice matters. Random forest records F1/AP of 0.914/0.978 with pooled gates, 0.636/0.851 with one netlist held out, and 0.460/0.577 with a host family held out. XGBoost falls from 0.946/0.976 to 0.464/0.544 across the same comparison. Logistic regression loses AP, although its fixed-threshold F1 is not monotonic. Each family shows the same pooled-to-family direction. Feature removal, repeated model and simulator seeds, score normalization, parser-related exclusions, and a smaller sample change the size of the gap without reversing it. Aggregation also matters: a gate-weighted average is dominated by the larger ISCAS files, so the headline values give each host family one vote. Bootstrap and jackknife summaries keep the gap positive, but their folds reuse training families. We treat the five family rows as descriptive evidence rather than independent trials. Five host families are too few for a population claim, and the experiment says nothing about transfer to a new cell library or an industrial design. It supports a narrower conclusion: sibling benchmark variants can inflate apparent transfer. Benchmarks with several variants of one host circuit should report family-aware holdouts and all five family results beside pooled scores.
| Comments: | 7 pages, 2 figures, 3 table. Accepted at IEEE ICCSIE 2026. Author accepted manuscript. Reproducibility artifact: this https URL |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computational Engineering, Finance, and Science (cs.CE); Databases (cs.DB); Computer Science and Game Theory (cs.GT) |
| Cite as: | arXiv:2609.07199 [cs.LG] |
| (or arXiv:2609.07199v2 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2609.07199 arXiv-issued DOI via DataCite |
Submission history
From: Kainan Zhou [view email]
[v1]
Mon, 7 Sep 2026 08:22:53 UTC (394 KB)
[v2]
Fri, 9 Oct 2026 04:56:55 UTC (394 KB)
来源：arXiv cs.AI · arxiv.org