Amit K. Verma
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Amit K. Verma has a total of 31 co-authors. Here are the top co-authors by number of shared publications:
- Anthony D. Rollett
Carnegie Mellon University
3 shared publications - Jennifer L. W. Carter
Department of Materials Science and Engineering, Case Western Reserve University, Cleveland, USA
3 shared publications - Emma Strubell
Carnegie Mellon University
2 shared publications - Junwon Seo
Materials Science and Engineering, Carnegie Mellon University, Pittsburgh, USA
2 shared publications - Laura S. Bruckman
Case Western Reserve University
2 shared publications - Robin Kuo
Materials Science and Engineering, Carnegie Mellon University, Pittsburgh, USA
2 shared publications
Latest Publications
- 26 September 2025
- Published by Springer Nature in Integrating Materials and Manufacturing Innovation
- Vol. 14 (4) , 643-656
Text Mining for Process–Structure–Properties Relationships in Metals
- 26 September 2025
- Published by Springer Nature in Integrating Materials and Manufacturing Innovation
- Vol. 14 (4) , 643-656
Text Mining for Process–Structure–Properties Relationships in Metals
- Open Access
Structured Extraction of Process Structure Properties Relationships in Materials Science
With the advent of large language models (LLMs), the vast unstructured text within millions of academic papers is increasingly accessible for materials discovery, although significant challenges remain. While LLMs offer promising few- and zero-shot learning capabilities, particularly valuable in the materials domain where expert annotations are scarce, general-purpose LLMs often fail to address key materials-specific queries without further adaptation. To bridge this gap, fine-tuning LLMs on human-labeled data is essential for effective structured knowledge extraction. In this study, we introduce a novel annotation schema designed to extract generic process-structure-properties relationships from scientific literature. We demonstrate the utility of this approach using a dataset of 128 abstracts, with annotations drawn from two distinct domains: high-temperature materials (Domain I) and uncertainty quantification in simulating materials microstructure (Domain II). Initially, we developed a conditional random field (CRF) model based on MatBERT, a domain-specific BERT variant, and evaluated its performance on Domain I. Subsequently, we compared this model with a fine-tuned LLM (GPT-4o from OpenAI) under identical conditions. Our results indicate that fine-tuning LLMs can significantly improve entity extraction performance over the BERT-CRF baseline on Domain I. However, when additional examples from Domain II were incorporated, the performance of the BERT-CRF model became comparable to that of the GPT-4o model. These findings underscore the potential of our schema for structured knowledge extraction and highlight the complementary strengths of both modeling approaches.
Open AccessStructured Extraction of Process Structure Properties Relationships in Materials Science
With the advent of large language models (LLMs), the vast unstructured text within millions of academic papers is increasingly accessible for materials discovery, although significant challenges remain. While LLMs offer promising few- and zero-shot learning capabilities, particularly valuable in the materials domain where expert annotations are scarce, general-purpose LLMs often fail to address key materials-specific queries without further adaptation. To bridge this gap, fine-tuning LLMs on human-labeled data is essential for effective structured knowledge extraction. In this study, we introduce a novel annotation schema designed to extract generic process-structure-properties relationships from scientific literature. We demonstrate the utility of this approach using a dataset of 128 abstracts, with annotations drawn from two distinct domains: high-temperature materials (Domain I) and uncertainty quantification in simulating materials microstructure (Domain II). Initially, we developed a conditional random field (CRF) model based on MatBERT, a domain-specific BERT variant, and evaluated its performance on Domain I. Subsequently, we compared this model with a fine-tuned LLM (GPT-4o from OpenAI) under identical conditions. Our results indicate that fine-tuning LLMs can significantly improve entity extraction performance over the BERT-CRF baseline on Domain I. However, when additional examples from Domain II were incorporated, the performance of the BERT-CRF model became comparable to that of the GPT-4o model. These findings underscore the potential of our schema for structured knowledge extraction and highlight the complementary strengths of both modeling approaches.
- Open Access
The Effect of Interlayer Delay on the Heat Accumulation, Microstructures, and Properties in Laser Hot Wire Directed Energy Deposition of Ti-6Al-4V Single-Wall
Laser hot wire directed energy deposition (LHW-DED) is a layer-by-layer additive manufacturing technique that permits the fabrication of large-scale Ti-6Al-4V (Ti64) components with a high deposition rate and has gained traction in the aerospace sector in recent years. However, one of the major challenges in LHW-DED Ti64 is heat accumulation, which affects the part quality, microstructure, and properties of as-built specimens. These issues require a comprehensive understanding of the layerwise heat-accumulation-driven process–structure–property relationship in as-deposited samples. In this study, a systematic investigation was performed by fabricating three Ti-6Al-4V single-wall specimens with distinct interlayer delays, i.e., 0, 120, and 300 s. The real-time acquisition of high-fidelity thermal data and high-resolution melt pool images were utilized to demonstrate a direct correlation between layerwise heat accumulation and melt pool dimensions. The results revealed that the maximum heat buildup temperature of the topmost layer decreased from 660 C to 263 C with an increase to a 300 s interlayer delay, allowing for better control of the melt pool dimensions, which then resulted in improved part accuracy. Furthermore, the investigation of the location-specific composition, microstructure, and mechanical properties demonstrated that heat buildup resulted in the coarsening of microstructures and, consequently, the reduction of micro-hardness with increasing height. Extending the delay by 120 s resulted in a 5% improvement in the mechanical properties, including an increase in the yield strength from 817 MPa to 859 MPa and the ultimate tensile strength from 914 MPa to 959 MPa. Cooling rates estimated at 900 C using a one-dimensional thermal model based on a numerical method allowed us to establish the process–structure–property relationship for the wall specimens. The study provides deeper insight into the effect of heat buildup in LHW-DED and serves as a guide for tailoring the properties of as-deposited specimens by regulating interlayer delay.Open AccessThe Effect of Interlayer Delay on the Heat Accumulation, Microstructures, and Properties in Laser Hot Wire Directed Energy Deposition of Ti-6Al-4V Single-Wall
Laser hot wire directed energy deposition (LHW-DED) is a layer-by-layer additive manufacturing technique that permits the fabrication of large-scale Ti-6Al-4V (Ti64) components with a high deposition rate and has gained traction in the aerospace sector in recent years. However, one of the major challenges in LHW-DED Ti64 is heat accumulation, which affects the part quality, microstructure, and properties of as-built specimens. These issues require a comprehensive understanding of the layerwise heat-accumulation-driven process–structure–property relationship in as-deposited samples. In this study, a systematic investigation was performed by fabricating three Ti-6Al-4V single-wall specimens with distinct interlayer delays, i.e., 0, 120, and 300 s. The real-time acquisition of high-fidelity thermal data and high-resolution melt pool images were utilized to demonstrate a direct correlation between layerwise heat accumulation and melt pool dimensions. The results revealed that the maximum heat buildup temperature of the topmost layer decreased from 660 C to 263 C with an increase to a 300 s interlayer delay, allowing for better control of the melt pool dimensions, which then resulted in improved part accuracy. Furthermore, the investigation of the location-specific composition, microstructure, and mechanical properties demonstrated that heat buildup resulted in the coarsening of microstructures and, consequently, the reduction of micro-hardness with increasing height. Extending the delay by 120 s resulted in a 5% improvement in the mechanical properties, including an increase in the yield strength from 817 MPa to 859 MPa and the ultimate tensile strength from 914 MPa to 959 MPa. Cooling rates estimated at 900 C using a one-dimensional thermal model based on a numerical method allowed us to establish the process–structure–property relationship for the wall specimens. The study provides deeper insight into the effect of heat buildup in LHW-DED and serves as a guide for tailoring the properties of as-deposited specimens by regulating interlayer delay. - 1 February 2020
- Published by Springer Nature in MRS Advances
- Vol. 5 (7) , 319-327
Harnessing Legacy Data to Educate Data-Enabled Structural Materials Engineers
- 1 February 2020
- Published by Springer Nature in MRS Advances
- Vol. 5 (7) , 319-327
Harnessing Legacy Data to Educate Data-Enabled Structural Materials Engineers
- 1 January 2020
- Published by Elsevier in Journal of Alloys and Compounds
- Vol. 815
Predictions of long-term creep life for the family of 9–12 wt% Cr martensitic steels
- 1 January 2020
- Published by Elsevier in Journal of Alloys and Compounds
- Vol. 815
Predictions of long-term creep life for the family of 9–12 wt% Cr martensitic steels
- 1 August 2019
- Published by Elsevier in Materials Science and Engineering: A
- Vol. 763, 138142
Screening of heritage data for improving toughness of creep-resistant martensitic steels
- 1 August 2019
- Published by Elsevier in Materials Science and Engineering: A
- Vol. 763, 138142
Screening of heritage data for improving toughness of creep-resistant martensitic steels
- 29 April 2019
- Published by Springer Nature in Metallurgical and Materials Transactions A
- Vol. 50 (7) , 3106-3120
Mapping Multivariate Influence of Alloying Elements on Creep Behavior for Design of New Martensitic Steels
- 29 April 2019
- Published by Springer Nature in Metallurgical and Materials Transactions A
- Vol. 50 (7) , 3106-3120
Mapping Multivariate Influence of Alloying Elements on Creep Behavior for Design of New Martensitic Steels