Accurate, Interdisciplinary and Transparent Structure-property Understanding with Deep Native Structural Reasoning
作者: Chen Tang, Yizhou Wang, Jianyu Wu, Lintao Wang, Shixiang Tang, Pengze Li, Encheng Su, Jun Yao, Jiabei Xiao, Yuqi Shi, Jielan Li, Hongxia Hao, Zhangyang Gao, Fang Wu, Ben Fei, Xiangyu Yue, Pan Tan, Bozitao Zhong, Jinouwen Zhang, Aoran Wang, Yan Lu, Jiaheng Liu, Xinzhu Ma, Liang Hong, Mingyue Zheng, Phil Torr, Bowen Zhou, Wanli Ouyang, Lei Bai
分类: cs.CL, cs.AI, cs.CE, cs.LG
发布日期: 2026-07-08
💡 一句话要点
提出SciReasoner以解决结构-属性关系理解问题
🎯 匹配领域: 支柱九：具身大模型 (Embodied Foundation Models)
关键词: 结构-属性关系 深度学习 多模态模型 科学推理 生物信息学 药物发现 材料科学
📋 核心要点
- 现有方法在结构-属性关系的理解上面临表示和推理的双重挑战，难以同时保留领域特有的结构信息和科学约束。
- 论文提出SciReasoner，通过将结构信息离散化为结构感知词汇，增强了模型在推理过程中的可解释性和准确性。
- SciReasoner在86个基准测试中实现了67项任务的最先进性能，尤其在低同源蛋白质的细胞组分注释和化学反应预测中显著提升。
📝 摘要（中文）
结构-属性关系是生物学、化学和材料科学的基础，功能、反应性和物理响应源于空间、化学和周期性组织。解释这些关系需要通过科学原理和物理约束来解读结构证据。本文提出SciReasoner，一个多模态科学基础模型，能够在蛋白质、小分子和无机晶体中进行本土结构推理。SciReasoner将坐标、拓扑和周期连接离散化为统一的结构感知词汇，在推理过程中将结构标记视为可寻址的证据单元。实验结果显示，SciReasoner在多个领域的表现优于现有方法，连接了准确预测与可解释的科学推理。
🔬 方法详解
问题定义：论文要解决的问题是如何有效地理解和解释结构-属性关系，现有方法在表示和推理过程中无法兼顾领域特有的结构信息和科学约束，导致预测准确性不足。
核心思路：论文的核心解决思路是引入SciReasoner模型，通过将结构信息离散化为统一的结构感知词汇，使得模型在推理时能够更好地利用结构证据。这样的设计旨在提高模型的可解释性和准确性。
技术框架：SciReasoner的整体架构包括多个模块，首先对坐标、拓扑和周期连接进行离散化，然后将这些结构标记作为可寻址的证据单元进行推理。模型通过多模态输入处理不同类型的结构数据。
关键创新：最重要的技术创新点在于将结构信息转化为可操作的证据单元，使得模型在推理过程中能够直接利用结构特征。这一方法与现有的基于文本的推理模型有本质区别。
关键设计：在模型设计中，SciReasoner采用了特定的损失函数来优化推理过程，并在网络结构上进行了调整，以适应多模态输入的特性。
🖼️ 关键图片
📊 实验亮点
在实验中，SciReasoner在低同源蛋白质的细胞组分注释中将F_max从0.42提升至0.55，在化学领域的单步逆合成准确率从0.63提升至0.72。此外，SciReasoner在86个基准测试中实现了67项任务的最先进性能，专家评估中98%的案例认为其推理过程优于现有的大型语言模型。
🎯 应用场景
SciReasoner的潜在应用领域包括生物信息学、药物发现和材料设计等。通过提高结构-属性关系的理解能力，该模型能够加速新材料的开发和新药的发现，具有重要的实际价值和未来影响。
📄 摘要（原文）
Structure-property relationships are foundational to biology, chemistry and materials science, where function, reactivity and physical response emerge from spatial, chemical and periodic organization. Mechanistically explaining these relationships requires interpreting structural evidence through scientific principles and physical constraints, from stereochemistry and bonding to symmetry, energetics and periodic order. However, applying artificial intelligence to this process presents a joint challenge of representation and reasoning: models must preserve domain-native structural information while showing how specific evidence supports predictions under these constraints. Here we introduce SciReasoner, a multimodal scientific foundation model for native structural reasoning across proteins, small molecules and inorganic crystals. SciReasoner discretizes coordinates, topologies and periodic connectivities into a unified structure-aware vocabulary, treating structural tokens as addressable evidence units during reasoning. In homology-controlled Gene Ontology prediction, SciReasoner improves Cellular Component annotation for low-homology and orphan-like proteins, increasing $F_{\max}$ from 0.42 to 0.55. In chemistry, it raises single-step retrosynthesis accuracy from 0.63 to 0.72 while generating fragment-level disconnection and precursor-verification traces. In materials science, its representations separate elemental and compound phases and resolve high- and low-band-gap regimes. Across 86 benchmarks, SciReasoner achieves state-of-the-art performance on 67 tasks. Double-blind expert evaluation rates its reasoning traces as preferred or at least comparable to those of a frontier large language model in 98% of cases. By making structure an inspectable substrate for reasoning under scientific constraints, SciReasoner connects accurate prediction with interpretable scientific inference.