arXiv cs.AI· Stepan N. Ostarkov, Gregorii A. Semakin, Ivan A. Bespalov, Michael G. Medvedev·· 11 小时前AI 评分58
通用基础模型在化学性质预测任务上超越专用化学模型
General-purpose foundation models surpass specialized ones on chemical property prediction tasks
AI 导读
研究比较了化学专用基础模型（ChemBERTa、MoLFormer）与通用基础模型（GPT-6 Astra、Claude Opus 5.5）在化学各领域任务上的准确率。
正文 · 原文
Abstract:Foundation models (FMs) learn reusable representations from large collections of unlabeled data. In chemistry the corresponding data mainly consists of molecular structures, which are abundant, whereas experimental measurements are scarce. Chemical FMs such as ChemBERTa and MoLFormer were therefore trained on diverse molecular structures to enable fine-tuning on small experimental datasets (less than 100 datapoints in most realistic scenarios) and subsequent property prediction. At the same time, general-purpose FMs such as GPT-6 Astra and Claude Opus 5.5 were trained on virtually all publicly available data, including knowledge of the humankind on molecular properties and laws governing them. Here we compare accuracies of chemical and general-purpose FMs on tasks from various areas of chemistry. Strikingly, general-purpose GPT-6 Astra outperforms the currently eligible chemical FMs on five physicochemical and quantum-chemical regression tasks when only 60 experimental datapoints are available. Only when provided 10x datapoints do the best chemical FMs reach performance of GPT-6 Astra: on 1,000 and 10,000 points they show performance on par with Astra on 60 and 1,000 points, respectively. Thus, counterintuitively, frontier general-purpose FMs are better suited for guiding chemical research in data-scarce areas than specialized chemical FMs, suggesting that learning relevant primitives (e.g. chemical structures) does not substitute training on relevant humankind knowledge.
| Subjects: | Chemical Physics (physics.chem-ph); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.41899 [physics.chem-ph] |
| (or arXiv:2610.41899v1 [physics.chem-ph] for this version) | |
| https://doi.org/10.48550/arXiv.2610.41899 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Michael G. Medvedev [view email]
[v1]
Fri, 30 Oct 2026 15:37:40 UTC (515 KB)
来源：arXiv cs.AI · arxiv.org