用双模态超球嵌入提升生物分类的层次结构建模效果
Hyperbolic Multimodal Representation Learning for Biological Taxonomies
- 将图像与基因数据联合嵌入超球空间，采用对比学习与新提出的层级蕴含目标
- 在未见物种分类任务中，基于DNA条形码的表现优于所有基线模型
- 适合做物种发现与生态监测的结构化建模，尤其对层次关系敏感的任务
生物多样性研究中的分类任务需依据多模态证据（如图像与基因信息）将生物标本组织成结构化层级。本文探究超球网络是否能为这类层次模型提供更优的嵌入空间。方法通过对比学习与一种新型堆叠蕴含目标，将多模态输入统一嵌入共享超球空间。在BIOSCAN-1M数据集上的实验表明，超球嵌入在性能上可媲美欧氏基线模型，并在使用DNA条形码的未见物种分类任务中表现最优。然而，细粒度分类与开放世界泛化仍具挑战。该框架为生物多样性建模提供了结构感知基础，有望应用于物种发现、生态监测与保护工作。
原文摘要 · Abstract (English)
Taxonomic classification in biodiversity research involves organizing biological specimens into structured hierarchies based on evidence, which can come from multiple modalities such as images and genetic information. We investigate whether hyperbolic networks can provide a better embedding space for such hierarchical models. Our method embeds multimodal inputs into a shared hyperbolic space using contrastive and a novel stacked entailment-based objective. Experiments on the BIOSCAN-1M dataset show that hyperbolic embedding achieves competitive performance with Euclidean baselines, and outperforms all other models on unseen species classification using DNA barcodes. However, fine-grained classification and open-world generalization remain challenging. Our framework offers a structure-aware foundation for biodiversity modelling, with potential applications to species discovery, ecological monitoring, and conservation efforts.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。