修复3D分子生成基准数据集的化学错误，提升评估准确性
GEOM-Drugs Revisited: Toward More Chemically Accurate Benchmarks for 3D Molecule Generation
- 修正数据预处理中的化合价与键级计算错误
- 基于GFN2-xTB构建更准确的几何与能量评估标准
- 提供可复现的评估工具，适合分子生成研究者使用
深度生成模型在生成有效3D分子结构方面展现出显著潜力，其中GEOM-Drugs数据集是关键基准。然而现有评估协议存在严重缺陷，包括错误的化合价定义、键级计算漏洞，以及与参考数据不一致的力场依赖。本文重新审视GEOM-Drugs，提出修正后的评估框架：识别并修复数据预处理问题，构建化学上准确的化合价表，并引入基于GFN2-xTB的几何与能量基准。我们在该框架下重新训练和评估多个领先模型，提供更新后的性能指标及未来基准测试的实用建议。结果强调了在3D分子生成中采用化学严谨评估的重要性。推荐的评估方法与GEOM-Drugs处理脚本已公开于https://github.com/isayevlab/geom-drugs-3dgen-evaluation。
原文摘要 · Abstract (English)
Deep generative models have shown significant promise in generating valid 3D molecular structures, with the GEOM-Drugs dataset serving as a key benchmark. However, current evaluation protocols suffer from critical flaws, including incorrect valency definitions, bugs in bond order calculations, and reliance on force fields inconsistent with the reference data. In this work, we revisit GEOM-Drugs and propose a corrected evaluation framework: we identify and fix issues in data preprocessing, construct chemically accurate valency tables, and introduce a GFN2-xTB-based geometry and energy benchmark. We retrain and re-evaluate several leading models under this framework, providing updated performance metrics and practical recommendations for future benchmarking. Our results underscore the need for chemically rigorous evaluation practices in 3D molecular generation. Our recommended evaluation methods and GEOM-Drugs processing scripts are available at https://github.com/isayevlab/geom-drugs-3dgen-evaluation.
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。