做有机合成路线规划的化学家终于有了一个既能给出准确条件推荐、又能展示具体先例的系统——HiRes 让你不再盲目信任黑箱模型，建议点开看看它是如何平衡精度与可解释性的。
HiRes 是一种检索增强的反应条件推荐系统，它结合了图编码器、变换感知交叉注意力、多流反应融合和 k-NN 检索层，在 USPTO-Condition 数据集上取得了催化剂、溶剂和试剂 top-1 准确率分别为 0.929、0.534 和 0.530 的最优结果。该系统不仅提供高精度预测，还能返回可检视的化学先例，帮助化学家理解推荐依据。与纯参数化方法相比，HiRes 在溶剂和试剂选择上通过配对自助法分析显示出统计显著的增益。HiRes 弥合了预测准确性与化学可解释性之间的差距，为实际合成规划提供了既准确又透明的推荐。
HiRes: Inspectable Precedent Memory for Reaction Condition Recommendation
Reaction condition recommendation sits immediately after retrosynthetic disconnection selection, and in practice, chemists require both accurate predictions and the precedents that justify them. We present HiRes (Hierarchical Reaction Representations), a retrieval-augmented condition recommendation system whose learned reaction space serves as both a classifier feature and an inspectable precedent memory. The model combines a graph encoder, transformation-aware cross-attention, multi-stream reaction fusion, and a k-NN retrieval layer. HiRes achieves state-of-the-art performance among primary-slot USPTO-Condition models, reaching Catalyst, Solvent, and Reagent top-1 accuracies (Acc@1) of 0.929, 0.534, and 0.530 respectively. It ties the best reported baseline on Catalyst while outperforming models such as REACON on Solvent and Reagent. Furthermore, paired bootstrap analysis demonstrates that integrating retrieval with learned condition heads provides statistically significant gains for solvent and reagent selection over purely parametric approaches. Ultimately, HiRes bridges the gap between predictive accuracy and chemical interpretability, offering a single representation that supplies both competitive recommendations and the concrete chemical precedents necessary for practical synthesis planning.