发表机构
Ludwig-Maximilians-Universität München; University of the Free State（慕尼黑大学; 自由州大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究提出利用模型生成的伪标签进行半监督学习，在欧洲18物种语料库及南非野外音频上验证其有效性，并引入属感知平滑提升识别性能，显著缩小与全监督的差距。
AI 中文摘要
被动声学监测产生的蝙蝠录音数量远超专家能够标注的范围。我们证明，简单的模型生成的伪标签能够将这些多余数据转化为有效的监督信号。我们在一个包含18个欧洲物种的语料库上，仅使用其10%的训练标签，将伪标签方法与其他半监督学习方法进行了比较，然后将表现最佳的方法迁移到包含九个蝙蝠分类单元和一个干扰类别的南非野外音频数据上。伪标签方法在欧洲数据集的所有评估指标上均优于其他半监督学习方法，恢复了与全监督方法差距的61.5%。该方法迁移到野外音频数据后，物种准确率提升了10.69个百分点，物种宏F1分数提升了4.96个百分点。我们还引入了属感知平滑，将不确定的目标概率质量导向同属物种。结合均匀平滑，该方法达到了79.16的物种宏F1分数，比硬目标高出4.73个百分点。因此，简单的伪标签在此生态数据规模上非常有效，而属感知目标在不增加标注成本的情况下注入了有用的生物学结构。此 https URL
英文摘要
Passive acoustic monitoring produces far more bat recordings than experts can label. We show that simple model-generated pseudo-labels turn this surplus into effective supervision. We compare pseudo-labeling with other semi-supervised learning methods on an 18-species European corpus using only 10% of its training labels, then transfer the strongest approaches to South African field audio containing nine bat taxa and a nuisance class. Pseudo-labeling outperforms the other semi-supervised learning methods on every European measure, recovering up to 61.5% of the gap to full supervision. It transfers to field audio with gains of 10.69 points in species accuracy and 4.96 points in species macro-F1. We also introduce genus-aware smoothing, which directs uncertain target mass toward congeneric species. Combined with uniform smoothing, it reaches 79.16 species macro-F1, 4.73 points above hard targets. Simple pseudo-labels are therefore highly effective at this ecological data scale, while genus-aware targets inject useful biological structure at no annotation cost. https://code4conservation.github.io/UnlabeledEchoes/
CommentsAccepted at CV4Ecology @ ECCV 2026
