Road of Large Language Model: Source, Challenge, and Future Perspectives.

Research (Wash D C)

School of Computer Science and Technology, Xidian University, Xi'an, China.

Published: July 2025


Category Ranking

98%

Total Visits

921

Avg Visit Duration

2 minutes

Citations

20

Article Abstract

Language model (LM), a foundational algorithm in the development of capable artificial intelligence, has been widely explored, achieving remarkable attainment. As research advances, large language models (LLMs) have emerged by pretraining transformer-based models on large-scale corpora. These models showed great zero-shot and few-shot learning capabilities across a variety of tasks, attracting widespread attention from both academia and industry. Despite impressive performance, LLMs still tackle challenges in addressing complex real-world scenarios. Recently, the advent of DeepSeek has reignited intense interest among researchers. In this paper, we provide a concise development history of LLM and discuss current challenges and future perspective. In practice, we focus on 4 crucial aspects of LLMs, including emergent abilities, human alignment, retrieval augmented generation, and applications in specific domains.

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC12304883PMC
http://dx.doi.org/10.34133/research.0655DOI Listing

Publication Analysis

Top Keywords

large language
8
language model
8
road large
4
model source
4
source challenge
4
challenge future
4
future perspectives
4
perspectives language
4
model foundational
4
foundational algorithm
4

Similar Publications

A plain language summary of the MIRACLE study: benralizumab in people in Asia with severe asthma.

Immunotherapy

September 2025

aGuangzhou Institute of Respiratory Health, State Key Laboratory of Respiratory Disease, National Clinical Research Center for Respiratory Disease, National Center for Respiratory Medicine, The First Affiliated Hospital of Guangzhou Medical University, Guangzhou, China.

View Article and Find Full Text PDF

Applications driven by large language models (LLMs) are reshaping higher education by offering innovative tools that enhance learning, streamline administrative tasks, and support scholarly work. However, their integration into education institutions raises ethical concerns related to bias, misinformation, and academic integrity, necessitating thoughtful institutional responses. This article explores the evolving role of LLMs in midwifery higher education, providing historical context, key capabilities, and ethical considerations.

View Article and Find Full Text PDF

A growing literature explores the representational detail of infants' early lexical representations, but no study has investigated how exposure to real-life acoustic-phonetic variation impacts these representations. Indeed, previous experimental work with young infants has largely ignored the impact of accent exposure on lexical development. We ask how routine exposure to accent variation affects 6-month-olds' ability to detect mispronunciations.

View Article and Find Full Text PDF

Objectives: The primary aim of this study was to compare resource utilization between lower and higher-risk brief resolved unexplained events (BRUE) in the general (GED) and pediatric (PED) emergency departments.

Methods: We conducted a retrospective chart review of BRUE cases from a large health system over 6-and-a-half years. Our primary outcome was the count of diagnostic tests per encounter.

View Article and Find Full Text PDF

Large language models (LLMs) have been successfully used for data extraction from free-text radiology reports. Most current studies were conducted with LLMs accessed via an application programming interface (API). We evaluated the feasibility of using open-source LLMs, deployed on limited local hardware resources for data extraction from free-text mammography reports, using a common data element (CDE)-based structure.

View Article and Find Full Text PDF