Search research articles

关于 JoVE

概览领导团队博客 JoVE 帮助中心

作者

出版流程编辑委员会范围与政策同行评审常见问题投稿

图书馆员

用户评价订阅访问资源图书馆顾问委员会常见问题

研究

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments 存档

教育

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual 教师资源中心教师网站

使用条款与条件

相关概念视频

Extraction: Partition and Distribution Coefficients

Extraction: Partition and Distribution Coefficients

The distribution law or Nernst's distribution law is the law that governs the distribution of a solute between two immiscible solvents. This law, also known as the partition law, states that if a solute is added to the mixture of two immiscible solvents at a constant temperature, the solute is distributed between the two solvents in such a way that the ratio of solute concentrations in the solvents remains constant at equilibrium.
For extracting a solute from an aqueous phase into an organic...

您也可能阅读

相关文章

通过共同作者、期刊和引用图与本文相关的文章。

排序

Same author

Enhanced Ionic Conductivity at the Solid Electrolyte Interphase of Oxygen-Doped Li<sub>6</sub>PS<sub>5</sub>Cl.

Advanced science (Weinheim, Baden-Wurttemberg, Germany)·2026

Same author

SPACESHIP: Autonomous Mapping of Hardware-Dependent Synthesizable Space in Solution-Phase Gold Nanomaterials.

Journal of the American Chemical Society·2026

Same author

Proteolytic cleavage of G3BP1 by calpain 1 couples NMDAR activation to mTOR-dependent local translation.

EMBO reports·2026

Same author

Toward self-driving laboratory 2.0 for chemistry and materials discovery.

Materials horizons·2026

Same author

Ligand-Driven Tuning of Adsorption Energy in Nanocrystals for High-Performance H<sub>2</sub>O<sub>2</sub> Electrosynthesis.

Journal of the American Chemical Society·2026

Same author

Accelerating High-Entropy Alloy Design via Machine Learning: Predicting Yield Strength from Composition.

Materials (Basel, Switzerland)·2026

Search research articles

相关实验视频

Updated: Jun 14, 2026

Atom Probe Tomography Studies on the CuIn,GaSe2 Grain Boundaries

Atom Probe Tomography Studies on the CuIn,GaSe2 Grain Boundaries

Published on: April 22, 2013

MaTableGPT:基于GPT的表格数据提取器,从材料科学文学中提取数据.

Gyeong Hoon Yi^1,2, Jiwoo Choi^1,2, Hyeongyun Song¹

¹Computational Science Research Center, Korea Institute of Science and Technology, Seoul, 02792, Republic of Korea.

Advanced science (Weinheim, Baden-Wurttemberg, Germany)

|January 24, 2025

概括

基于GPT的工具MaTableGPT有效地从各种材料科学表中提取数据. 它实现了水分裂催化数据的高准确性,用几次射击学习提供了成本和性能的最佳平衡.

关键词:

在 GPT 中,GPT 必须是 GPT.大型语言模型.文学采矿采矿文学采矿机器学习是机器学习.材料科学材料科学材料科学表格数据提取表格数据提取水的分裂催化剂.

更多相关视频

Detection and Recovery of Palladium, Gold and Cobalt Metals from the Urban Mine Using Novel Sensors/Adsorbents Designated with Nanoscale Wagon-wheel-shaped Pores

Detection and Recovery of Palladium, Gold and Cobalt Metals from the Urban Mine Using Novel Sensors/Adsorbents Designated with Nanoscale Wagon-wheel-shaped Pores

Published on: December 6, 2015

Determining the Mechanical Strength of Ultra-Fine-Grained Metals

Determining the Mechanical Strength of Ultra-Fine-Grained Metals

Published on: November 22, 2021

相关实验视频

Last Updated: Jun 14, 2026

Atom Probe Tomography Studies on the CuIn,GaSe2 Grain Boundaries

Atom Probe Tomography Studies on the CuIn,GaSe2 Grain Boundaries

Published on: April 22, 2013

Detection and Recovery of Palladium, Gold and Cobalt Metals from the Urban Mine Using Novel Sensors/Adsorbents Designated with Nanoscale Wagon-wheel-shaped Pores

Detection and Recovery of Palladium, Gold and Cobalt Metals from the Urban Mine Using Novel Sensors/Adsorbents Designated with Nanoscale Wagon-wheel-shaped Pores

Published on: December 6, 2015

Determining the Mechanical Strength of Ultra-Fine-Grained Metals

Determining the Mechanical Strength of Ultra-Fine-Grained Metals

Published on: November 22, 2021

科学领域:

材料科学材料科学材料科学
计算化学计算化学
数据科学数据科学数据科学

背景情况:

从科学文献中的各种表格中提取数据对于创建数据库至关重要.
现有的基于规则的方法是无效的,因为材料科学论文中的表格格式不同.

研究的目的:

开发MaTableGPT,这是一个基于GPT的系统,用于从材料科学文献中提取表格数据.
为了提高GPT的理解和过幻觉信息,以获得准确的数据提取.

主要方法:

使用基于GPT的架构,并采用专门的策略来表格数据表示和分割.
实施了后续问题机制,以过幻觉信息.
评估了零射击,少数射击和微调学习方法,以提高成本效益和准确性.

主要成果:

在水分裂催化学文献中,MaTableGPT实现了高提取精度 (高达96.8%的F1得分).
少数射击学习证明了高精度 (>95%F1得分) 和低成本 (GPT使用:5.97美元,标签:10个例子) 之间的最佳平衡.
对生成的数据库的统计分析提供了对催化剂属性的见解,例如过度潜在和元素利用.

结论:

MaTableGPT提供了一种有效的解决方案,可以从复杂的材料科学表中提取结构化数据.
短暂的学习提供了一个具有成本效益和准确的方法,用于使用大型语言模型提取数据.
生成的数据库有助于更深入地了解用于水分裂催化物的材料.