Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2510.13916 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918162364104704 |
|---|---|
| author | Li, Yuanhao Lai, Keyuan Wang, Tianqi Liu, Qihao Ma, Jiawei Hu, Yuan-Chao |
| author_facet | Li, Yuanhao Lai, Keyuan Wang, Tianqi Liu, Qihao Ma, Jiawei Hu, Yuan-Chao |
| contents | Accurate property data for chemical elements is crucial for materials design and manufacturing, but many of them are difficult to measure directly due to equipment constraints. While traditional methods use the properties of other elements or related properties for prediction via numerical analyses, they often fail to model complex relationships. After all, not all characteristics can be represented as scalars. Recent efforts have been made to explore advanced AI tools such as language models for property estimation, but they still suffer from hallucinations and a lack of interpretability. In this paper, we investigate Element2Vecto effectively represent chemical elements from natural languages to support research in the natural sciences. Given the text parsed from Wikipedia pages, we use language models to generate both a single general-purpose embedding (Global) and a set of attribute-highlighted vectors (Local). Despite the complicated relationship across elements, the computational challenges also exist because of 1) the discrepancy in text distribution between common descriptions and specialized scientific texts, and 2) the extremely limited data, i.e., with only 118 known elements, data for specific properties is often highly sparse and incomplete. Thus, we also design a test-time training method based on self-attention to mitigate the prediction error caused by Vanilla regression clearly. We hope this work could pave the way for advancing AI-driven discovery in materials science. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_13916 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Element2Vec: Build Chemical Element Representation from Text for Property Prediction Li, Yuanhao Lai, Keyuan Wang, Tianqi Liu, Qihao Ma, Jiawei Hu, Yuan-Chao Computation and Language Accurate property data for chemical elements is crucial for materials design and manufacturing, but many of them are difficult to measure directly due to equipment constraints. While traditional methods use the properties of other elements or related properties for prediction via numerical analyses, they often fail to model complex relationships. After all, not all characteristics can be represented as scalars. Recent efforts have been made to explore advanced AI tools such as language models for property estimation, but they still suffer from hallucinations and a lack of interpretability. In this paper, we investigate Element2Vecto effectively represent chemical elements from natural languages to support research in the natural sciences. Given the text parsed from Wikipedia pages, we use language models to generate both a single general-purpose embedding (Global) and a set of attribute-highlighted vectors (Local). Despite the complicated relationship across elements, the computational challenges also exist because of 1) the discrepancy in text distribution between common descriptions and specialized scientific texts, and 2) the extremely limited data, i.e., with only 118 known elements, data for specific properties is often highly sparse and incomplete. Thus, we also design a test-time training method based on self-attention to mitigate the prediction error caused by Vanilla regression clearly. We hope this work could pave the way for advancing AI-driven discovery in materials science. |
| title | Element2Vec: Build Chemical Element Representation from Text for Property Prediction |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2510.13916 |