提出安全多任务学习框架，提升挥发性与嗅觉阈值预测的泛化能力。
Safe Multitask Molecular Graph Networks for Vapor Pressure and Odor Threshold Prediction
- 以主任务为挥发性，辅任务为嗅觉阈值，采用延迟激活+梯度裁剪策略。
- 在骨架划分下，挥发性预测验证集均方误差达0.21（归一化空间）。
- 适合需高鲁棒性的分子属性预测研究者，尤其关注数据噪声场景。
本文研究气味相关属性建模中的两项重要任务：挥发性（VP）和嗅觉阈值（OP）。为评估模型在分布外（OOD）情况下的性能，采用Bemis-Murcko骨架划分方式。特征方面，引入丰富的A20/E17分子图特征（20维原子特征+17维键特征），并系统比较GINE与PNA骨干网络。结果显示：对于挥发性预测，使用PNA搭配简单回归头时，验证集均方误差约为0.21（归一化空间）；在相同骨架划分下，单任务嗅觉阈值预测通过稳健训练（Huber/winsor）实现验证集均方误差约0.60-0.61。针对多任务训练，提出“安全多任务”方法：以挥发性为主任务，嗅觉阈值为辅助任务，结合延迟激活、梯度裁剪和小权重设置，避免损害主任务表现，同时获得最优的挥发性泛化性能。论文提供完整可复现实验、消融研究及误差相似性分析，并讨论数据噪声影响与方法局限性。
原文摘要 · Abstract (English)
We investigate two important tasks in odor-related property modeling: Vapor Pressure (VP) and Odor Threshold (OP). To evaluate the model's out-of-distribution (OOD) capability, we adopt the Bemis-Murcko scaffold split. In terms of features, we introduce the rich A20/E17 molecular graph features (20-dimensional atom features + 17-dimensional bond features) and systematically compare GINE and PNA backbones. The results show: for VP, PNA with a simple regression head achieves Val MSE $\approx$ 0.21 (normalized space); for the OP single task under the same scaffold split, using A20/E17 with robust training (Huber/winsor) achieves Val MSE $\approx$ 0.60-0.61. For multitask training, we propose a **"safe multitask"** approach: VP as the primary task and OP as the auxiliary task, using delayed activation + gradient clipping + small weight, which avoids harming the primary task and simultaneously yields the best VP generalization performance. This paper provides complete reproducible experiments, ablation studies, and error-similarity analysis while discussing the impact of data noise and method limitations.
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。