Saved in:
Bibliographic Details
Main Authors: Wang, Suqing, Li, Zuchao, Shi, Luohe, Du, Bo, Zhao, Hai, Li, Yun, Wang, Qianren
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2509.18136
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915506906202112
author Wang, Suqing
Li, Zuchao
Shi, Luohe
Du, Bo
Zhao, Hai
Li, Yun
Wang, Qianren
author_facet Wang, Suqing
Li, Zuchao
Shi, Luohe
Du, Bo
Zhao, Hai
Li, Yun
Wang, Qianren
contents Large language models (LLMs) have achieved remarkable success across various domains, driving significant technological advancements and innovations. Despite the rapid growth in model scale and capability, systematic, data-driven research on how structural configurations affect performance remains scarce. To address this gap, we present a large-scale dataset encompassing diverse open-source LLM structures and their performance across multiple benchmarks. Leveraging this dataset, we conduct a systematic, data mining-driven analysis to validate and quantify the relationship between structural configurations and performance. Our study begins with a review of the historical development of LLMs and an exploration of potential future trends. We then analyze how various structural choices impact performance across benchmarks and further corroborate our findings using mechanistic interpretability techniques. By providing data-driven insights into LLM optimization, our work aims to guide the targeted development and application of future models. We will release our dataset at https://huggingface.co/datasets/DX0369/LLM-Structure-Performance-Dataset
format Preprint
id arxiv_https___arxiv_org_abs_2509_18136
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle From Parameters to Performance: A Data-Driven Study on LLM Structure and Development
Wang, Suqing
Li, Zuchao
Shi, Luohe
Du, Bo
Zhao, Hai
Li, Yun
Wang, Qianren
Machine Learning
Artificial Intelligence
Large language models (LLMs) have achieved remarkable success across various domains, driving significant technological advancements and innovations. Despite the rapid growth in model scale and capability, systematic, data-driven research on how structural configurations affect performance remains scarce. To address this gap, we present a large-scale dataset encompassing diverse open-source LLM structures and their performance across multiple benchmarks. Leveraging this dataset, we conduct a systematic, data mining-driven analysis to validate and quantify the relationship between structural configurations and performance. Our study begins with a review of the historical development of LLMs and an exploration of potential future trends. We then analyze how various structural choices impact performance across benchmarks and further corroborate our findings using mechanistic interpretability techniques. By providing data-driven insights into LLM optimization, our work aims to guide the targeted development and application of future models. We will release our dataset at https://huggingface.co/datasets/DX0369/LLM-Structure-Performance-Dataset
title From Parameters to Performance: A Data-Driven Study on LLM Structure and Development
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2509.18136