arXiv：cs.LG· Long Wang·· 4 小时前
通过受控类内结构变化构建决策源以提升共识伪标签学习
Constructing Structured Decision Sources for Consensus-Based Pseudo-Label Learning
AI 导读
研究者提出通过受控改变类内结构来构建决策源：从共享图表示出发调整中心粒度和邻域混合，复现每个源以检验稳定性，再用节点对共分配筛选互补子集并对一致预测排序用于学生训练。
正文
Abstract:Consensus can make pseudo-label learning more reliable, but only when its predictors contribute genuinely different evidence. Multiple models that repeat the same boundary provide additional votes without additional information. We address this problem by con structing decision sources through controlled changes to within-class structure. Starting from a shared graph representation, we vary center granularity and neighborhood mixing, reproduce each resulting source to test its stability, and select a complementary subset using node pair coassignment. Unanimous predictions from the selected sources are then ranked for student training. On the public fixed splits of Cora, CiteSeer, and PubMed, evaluated with five random seeds, the constructed sources improve fixed-budget training pseudo-label precision by 1.19 to 4.39 percentage points over three conventionally initialized GCN sources. Under matched structural filters, three-source consensus is more precise than each constituent source in all 45 dataset slot eed comparisons. The gains are strongest in pseudo-label quality: downstream accuracy remains competitive but does not lead on every dataset. These results identify source construction rather than model count alone as an important design problem for consensus-based pseudo-label learning.
| Comments: | 14 pages, 4 figures, 6 tables |
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.11621 [cs.LG] |
| (or arXiv:2610.11621v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.11621 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Long Wang [view email]
[v1]
Thu, 8 Oct 2026 10:00:26 UTC (3,082 KB)
来源：arXiv：cs.LG · arxiv.org