AMGPT: A large language model for contextual querying in additive manufacturing

IF 4.2 Q2 ENGINEERING, MANUFACTURING Additive manufacturing letters Pub Date : 2024-08-22 DOI:10.1016/j.addlet.2024.100232
Achuth Chandrasekhar , Jonathan Chan , Francis Ogoke , Olabode Ajenifujah , Amir Barati Farimani
{"title":"AMGPT: A large language model for contextual querying in additive manufacturing","authors":"Achuth Chandrasekhar ,&nbsp;Jonathan Chan ,&nbsp;Francis Ogoke ,&nbsp;Olabode Ajenifujah ,&nbsp;Amir Barati Farimani","doi":"10.1016/j.addlet.2024.100232","DOIUrl":null,"url":null,"abstract":"<div><p>Generalized large language models (LLMs) such as GPT-4 may not provide specific answers to queries formulated by materials science researchers. These models may produce a high-level outline but lack the capacity to return detailed instructions on manufacturing and material properties of novel alloys. We introduce “AMGPT”, a specialized LLM text generator designed for metal AM queries. The goal of AMGPT is to assist researchers and users in navigating a curated corpus of literature. Instead of training from scratch, we employ a pre-trained Llama2-7B model from Hugging Face in a Retrieval-Augmented Generation (RAG) setup, utilizing it to dynamically incorporate information from <span><math><mo>∼</mo></math></span>50 AM papers and textbooks in PDF format. Mathpix is used to convert these PDF documents into TeX format, facilitating their integration into the RAG pipeline managed by LlamaIndex. A query retrieval function has also been added, enabling the system to fetch relevant literature from Elsevier journals based on the context of the query. Expert evaluations of this project highlight that specific embeddings from the RAG setup accelerate response times and maintain coherence in the generated text.</p></div>","PeriodicalId":72068,"journal":{"name":"Additive manufacturing letters","volume":"11 ","pages":"Article 100232"},"PeriodicalIF":4.2000,"publicationDate":"2024-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S2772369024000409/pdfft?md5=8d7e38c2365561cad4541597909ff24b&pid=1-s2.0-S2772369024000409-main.pdf","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Additive manufacturing letters","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2772369024000409","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENGINEERING, MANUFACTURING","Score":null,"Total":0}
引用次数: 0

Abstract

Generalized large language models (LLMs) such as GPT-4 may not provide specific answers to queries formulated by materials science researchers. These models may produce a high-level outline but lack the capacity to return detailed instructions on manufacturing and material properties of novel alloys. We introduce “AMGPT”, a specialized LLM text generator designed for metal AM queries. The goal of AMGPT is to assist researchers and users in navigating a curated corpus of literature. Instead of training from scratch, we employ a pre-trained Llama2-7B model from Hugging Face in a Retrieval-Augmented Generation (RAG) setup, utilizing it to dynamically incorporate information from 50 AM papers and textbooks in PDF format. Mathpix is used to convert these PDF documents into TeX format, facilitating their integration into the RAG pipeline managed by LlamaIndex. A query retrieval function has also been added, enabling the system to fetch relevant literature from Elsevier journals based on the context of the query. Expert evaluations of this project highlight that specific embeddings from the RAG setup accelerate response times and maintain coherence in the generated text.

查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
AMGPT:用于增材制造语境查询的大型语言模型
通用大型语言模型(LLM),如 GPT-4,可能无法为材料科学研究人员提出的查询提供具体答案。这些模型可以生成一个高级大纲,但缺乏返回有关新型合金的制造和材料特性的详细说明的能力。我们介绍了 "AMGPT",这是一种专门为金属 AM 查询设计的 LLM 文本生成器。AMGPT 的目标是帮助研究人员和用户浏览经过整理的文献语料库。我们没有从头开始训练,而是在检索增强生成(RAG)设置中使用了来自 Hugging Face 的预训练 Llama2-7B 模型,并利用它动态纳入了来自 ∼50 篇 AM 论文和 PDF 格式教科书的信息。Mathpix 用于将这些 PDF 文档转换为 TeX 格式,便于将其整合到由 LlamaIndex 管理的 RAG 管道中。系统还增加了查询检索功能,可根据查询内容从爱思唯尔期刊中获取相关文献。该项目的专家评估强调,RAG 设置中的特定嵌入可加快响应时间,并保持生成文本的一致性。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
Additive manufacturing letters
Additive manufacturing letters Materials Science (General), Industrial and Manufacturing Engineering, Mechanics of Materials
CiteScore
3.70
自引率
0.00%
发文量
0
审稿时长
37 days
期刊最新文献
A new process route for the additive manufacturing of a high nitrogen containing martensitic stainless steel - A feasibility study Additive manufacturing simulations: An approach based on space partitioning and dynamic 3D mesh adaptation Understanding the effect of pre-sintering scanning strategy on the relative density of Zr-modified Al7075 processed by laser powder bed fusion Mechanical performance of laser powder bed fused Ti-6Al-4V: The influence of filter condition and part location Area-based composition predictions of materials fabricated using simultaneous wire-powder-directed energy deposition
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1