---
title: "SciReasoner：面向蛋白质、小分子和无机晶体的多模态科学推理模型"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2607.07708v1"
permalink: "https://feed.astralor.com/items/itm_6dcaed6fe13a022b31dee10e"
published: "2026-07-08T17:59:59.000Z"
score: 81
channel: "ai"
tags: ["SciReasoner","可解释AI","多模态科学基础模型","结构推理","跨学科"]
---

# SciReasoner：面向蛋白质、小分子和无机晶体的多模态科学推理模型

## AI 点评

> 这篇工作让我眼前一亮——它真正尝试解决AI在科学领域‘黑箱’的痛点，通过将结构token化为可推理的证据，把准确性提升和可解释性结合在了一起。对从事计算化学、结构生物学或材料基因组学的研究者来说，SciReasoner的跨学科能力和专家验证结果是重要参考。值得关注的是，作者已开源模型和推理轨迹，生态建设值得期待。

## 速览

该论文提出SciReasoner，一种多模态科学基础模型，能够对蛋白质、小分子和无机晶体进行本征结构推理。通过将坐标、拓扑和周期连接离散化为统一的结构感知词汇表，模型将结构token作为可寻址的推理证据单元。在跨学科基准测试中，SciReasoner在86项任务中达到67项最优，例如在低同源性蛋白质的基因本体预测中将Fmax从0.42提升至0.55，化学逆合成准确率从0.63提升至0.72。双盲专家评估表明，其推理轨迹在98%案例中优于或媲美前沿大语言模型。

## 关键点

- SciReasoner将空间坐标、拓扑和周期连接转换为统一的结构感知离散token，使模型能将结构作为可寻址证据进行推理。
- 在生物学、化学和材料科学86项基准测试中，SciReasoner在67项任务上达到最优，涵盖基因本体预测、逆合成分析和能带间隙分类。
- 双盲专家评估显示，SciReasoner的推理轨迹在98%的案例中优于或至少媲美前沿大语言模型。

## 全文

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.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_6dcaed6fe13a022b31dee10e