arXiv cs.CL· Harsidak Singh Banwait, Chander Prabha·· 12 小时前AI 评分20
从 Transformer 到 GPT-5：大语言模型设计的进展、局限与未来方向
Transformer to GPT-5: Advancements, Limitations, and Future Directions in Large Language Model Design
AI 导读
一篇综述按时间线梳理了大语言模型从 BERT 到 GPT、Gemini、Claude、Grok 的分类演进，覆盖训练数据集、算力、硬件规格、架构与参数量。研究指出，在架构复杂度与数据质量之间取得更优比例的模型，推理稳定性与对齐表现更强，且不会同比推高计算负载。
正文
Abstract:Large Language Models (LLMs) have revolutionized Artificial Intelligence (AI) by their extraordinary capabilities, like contextual understanding, few-shot/zero-shot learning, code generation, ability to reason and problem-solving. Recent studies have shown immense work on LLMs. Yet, there is limited work on the thorough taxonomy of LLMs. This study would showcase a detailed classification of LLMs, starting with the earlier models like BERT and leading to LLMs like GPT, Gemini, Claude and Grok. We would delve into topics like datasets used to train LLMs, computational power used during training, their hardware specifications, architecture and parameters. Further, we would also look at the limitations, computational challenges and open problems as well. This study would primarily focus on the timeline of LLMs, signifying their growth over the period. Findings reveal a clear pattern: models that optimize the ratio between architectural complexity and data quality exhibit stronger reasoning stability and better alignment, without proportionally increasing the computational load. This desired balance between scaling and sustainability is the defining characteristic of the current generation of AI.
| Comments: | 5 pages, ICIPCN |
| Subjects: | Computation and Language (cs.CL); Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.34281 [cs.CL] |
| (or arXiv:2610.34281v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.34281 arXiv-issued DOI via DataCite (pending registration) |
|
| Journal reference: | Published at 2026 6th International Conference on Image Processing and Capsule Networks (ICIPCN) |
Submission history
From: Harsidak Singh Banwait [view email]
[v1]
Sun, 25 Oct 2026 13:25:30 UTC (333 KB)
来源：arXiv cs.CL · arxiv.org