通过课程学习提升文本与分子结构检索效率与效果
CLASS: Enhancing Cross-Modal Text-Molecule Retrieval Performance and Training Efficiency
- 按难易程度动态调度样本，早期用简单样本加速训练
- 在ChEBI-20数据集上实现性能提升与训练时间显著减少
- 可适配任意主干模型，适合跨模态分子检索任务
跨模态文本-分子检索任务连接分子结构与自然语言描述。现有方法多聚焦于对齐文本与分子模态，却忽略了在不同训练阶段自适应调整学习状态及提升训练效率。为此，本文提出基于课程学习的跨模态文本-分子训练框架（CLASS），可与任意主干模型结合以获得显著性能提升。具体地，我们综合考虑文本与分子模态的样本难度，并设计样本调度器，按由易到难的顺序引入训练样本，显著减少训练初期的样本规模，提升训练效率；此外，引入自适应强度学习机制，随训练进程动态调节学习强度，统一控制各课程阶段的学习力度。在ChEBI-20数据集上的实验表明，所提方法不仅性能更优，且实现显著的时间节省。
原文摘要 · Abstract (English)
Cross-modal text-molecule retrieval task bridges molecule structures and natural language descriptions. Existing methods predominantly focus on aligning text modality and molecule modality, yet they overlook adaptively adjusting the learning states at different training stages and enhancing training efficiency. To tackle these challenges, this paper proposes a Curriculum Learning-bAsed croSS-modal text-molecule training framework (CLASS), which can be integrated with any backbone to yield promising performance improvement. Specifically, we quantify the sample difficulty considering both text modality and molecule modality, and design a sample scheduler to introduce training samples via an easy-to-difficult paradigm as the training advances, remarkably reducing the scale of training samples at the early stage of training and improving training efficiency. Moreover, we introduce adaptive intensity learning to increase the training intensity as the training progresses, which adaptively controls the learning intensity across all curriculum stages. Experimental results on the ChEBI-20 dataset demonstrate that our proposed method gains superior performance, simultaneously achieving prominent time savings.
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。