_version_ 1866911316036288512
author von Laszewski, Gregor
Brewer, Wesley
Thiyagalingam, Jeyan
Papay, Juri
Foundjem, Armstrong
Luszczek, Piotr
Emani, Murali
Moore, Shirley V.
Reddi, Vijay Janapa
Sinclair, Matthew D.
Lobentanzer, Sebastian
Goswami, Sujata
Hawks, Benjamin
Colombo, Marco
Tran, Nhan
Kirkpatrick, Christine R.
Alsudais, Abdulkareem
Barrett, Gregg
Li, Tianhao
Morehouse, Kirsten
Venkataraman, Shivaram
Jain, Rutwik
Mathur, Kartik
Lu, Victor
Singh, Tejinder
Mirza, Khojasteh Z.
Chen, Kongtao
Kunapuli, Sasidhar
Farrell, Gavin
Umeton, Renato
Fox, Geoffrey C.
author_facet von Laszewski, Gregor
Brewer, Wesley
Thiyagalingam, Jeyan
Papay, Juri
Foundjem, Armstrong
Luszczek, Piotr
Emani, Murali
Moore, Shirley V.
Reddi, Vijay Janapa
Sinclair, Matthew D.
Lobentanzer, Sebastian
Goswami, Sujata
Hawks, Benjamin
Colombo, Marco
Tran, Nhan
Kirkpatrick, Christine R.
Alsudais, Abdulkareem
Barrett, Gregg
Li, Tianhao
Morehouse, Kirsten
Venkataraman, Shivaram
Jain, Rutwik
Mathur, Kartik
Lu, Victor
Singh, Tejinder
Mirza, Khojasteh Z.
Chen, Kongtao
Kunapuli, Sasidhar
Farrell, Gavin
Umeton, Renato
Fox, Geoffrey C.
contents Benchmarks are a cornerstone of modern machine learning, enabling reproducibility, comparison, and scientific progress. However, AI benchmarks are increasingly complex, requiring dynamic, AI-focused workflows. Rapid evolution in model architectures, scale, datasets, and deployment contexts makes evaluation a moving target. Large language models often memorize static benchmarks, causing a gap between benchmark results and real-world performance. Beyond traditional static benchmarks, continuous adaptive benchmarking frameworks are needed to align scientific assessment with deployment risks. This calls for skills and education in AI Benchmark Carpentry. From our experience with MLCommons, educational initiatives, and programs like the DOE's Trillion Parameter Consortium, key barriers include high resource demands, limited access to specialized hardware, lack of benchmark design expertise, and uncertainty in relating results to application domains. Current benchmarks often emphasize peak performance on top-tier hardware, offering limited guidance for diverse, real-world scenarios. Benchmarking must become dynamic, incorporating evolving models, updated data, and heterogeneous platforms while maintaining transparency, reproducibility, and interpretability. Democratization requires both technical innovation and systematic education across levels, building sustained expertise in benchmark design and use. Benchmarks should support application-relevant comparisons, enabling informed, context-sensitive decisions. Dynamic, inclusive benchmarking will ensure evaluation keeps pace with AI evolution and supports responsible, reproducible, and accessible AI deployment. Community efforts can provide a foundation for AI Benchmark Carpentry.
format Preprint
id arxiv_https___arxiv_org_abs_2512_11588
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle AI Benchmark Democratization and Carpentry
von Laszewski, Gregor
Brewer, Wesley
Thiyagalingam, Jeyan
Papay, Juri
Foundjem, Armstrong
Luszczek, Piotr
Emani, Murali
Moore, Shirley V.
Reddi, Vijay Janapa
Sinclair, Matthew D.
Lobentanzer, Sebastian
Goswami, Sujata
Hawks, Benjamin
Colombo, Marco
Tran, Nhan
Kirkpatrick, Christine R.
Alsudais, Abdulkareem
Barrett, Gregg
Li, Tianhao
Morehouse, Kirsten
Venkataraman, Shivaram
Jain, Rutwik
Mathur, Kartik
Lu, Victor
Singh, Tejinder
Mirza, Khojasteh Z.
Chen, Kongtao
Kunapuli, Sasidhar
Farrell, Gavin
Umeton, Renato
Fox, Geoffrey C.
Artificial Intelligence
I.2.6
Benchmarks are a cornerstone of modern machine learning, enabling reproducibility, comparison, and scientific progress. However, AI benchmarks are increasingly complex, requiring dynamic, AI-focused workflows. Rapid evolution in model architectures, scale, datasets, and deployment contexts makes evaluation a moving target. Large language models often memorize static benchmarks, causing a gap between benchmark results and real-world performance. Beyond traditional static benchmarks, continuous adaptive benchmarking frameworks are needed to align scientific assessment with deployment risks. This calls for skills and education in AI Benchmark Carpentry. From our experience with MLCommons, educational initiatives, and programs like the DOE's Trillion Parameter Consortium, key barriers include high resource demands, limited access to specialized hardware, lack of benchmark design expertise, and uncertainty in relating results to application domains. Current benchmarks often emphasize peak performance on top-tier hardware, offering limited guidance for diverse, real-world scenarios. Benchmarking must become dynamic, incorporating evolving models, updated data, and heterogeneous platforms while maintaining transparency, reproducibility, and interpretability. Democratization requires both technical innovation and systematic education across levels, building sustained expertise in benchmark design and use. Benchmarks should support application-relevant comparisons, enabling informed, context-sensitive decisions. Dynamic, inclusive benchmarking will ensure evaluation keeps pace with AI evolution and supports responsible, reproducible, and accessible AI deployment. Community efforts can provide a foundation for AI Benchmark Carpentry.
title AI Benchmark Democratization and Carpentry
topic Artificial Intelligence
I.2.6
url https://arxiv.org/abs/2512.11588