EUVIMEDEUVIMED entdecken
Open Access · CC BY-ND 4.0

Transformer-Driven Large Language Models for Context-Aware Semantic Reasoning and Domain-Specific Text Generation

Edvinas Yamashiro

International Journal of Recent Advances in Engineering and Technology · 25.12.2025 · 14 (2) · 428-437

DOI: 10.65521/intjournalrecadvengtech.v14i2.2710 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Recent Advances in Engineering and Technology, online erschienen am 25.12.2025. Verfasst von Edvinas Yamashiro. Der Volltext ist frei zugänglich (Open Access, CC BY-ND 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Transformer-driven large language models (LLMs) have fundamentally transformed natural language processing by enabling context-aware semantic reasoning and high-quality text generation across diverse domains. This research proposes a comprehensive framework for leveraging transformer-based architectures to enhance contextual understanding and domain-specific text generation. The study focuses on integrating attention mechanisms, domain adaptation strategies, and fine-tuning techniques to improve semantic coherence and reasoning capabilities. The proposed approach utilizes pre-trained transformer models and adapts them through domain-specific fine-tuning and prompt optimization to generate accurate and contextually relevant outputs. Experimental evaluation demonstrates that transformer-based models outperform traditional sequence models in capturing long-range dependencies and generating coherent text. Additionally, domain adaptation techniques significantly improve performance in specialized applications such as healthcare, legal analysis, and technical writing. The study further investigates optimization strategies including reinforcement learning from human feedback (RLHF), retrieval-augmented generation, and parameter-efficient fine-tuning to enhance model efficiency and reliability. Results indicate that the proposed framework achieves superior performance in semantic reasoning tasks while maintaining scalability. This research contributes a structured methodology for designing context-aware LLM systems capable of generating high-quality domain-specific content.

Zitieren

APA

Yamashiro, E. (2025). Transformer-Driven Large Language Models for Context-Aware Semantic Reasoning and Domain-Specific Text Generation. International Journal of Recent Advances in Engineering and Technology, 14(2), 428-437. https://doi.org/10.65521/intjournalrecadvengtech.v14i2.2710

Vancouver

Yamashiro E. Transformer-Driven Large Language Models for Context-Aware Semantic Reasoning and Domain-Specific Text Generation. International Journal of Recent Advances in Engineering and Technology. 2025;14(2):428-437. doi:10.65521/intjournalrecadvengtech.v14i2.2710

BibTeX
@article{Yamashiro2025,
  author = {Yamashiro, Edvinas},
  title = {Transformer-Driven Large Language Models for Context-Aware Semantic Reasoning and Domain-Specific Text Generation},
  journal = {International Journal of Recent Advances in Engineering and Technology},
  year = {2025},
  volume = {14},
  number = {2},
  pages = {428--437},
  doi = {10.65521/intjournalrecadvengtech.v14i2.2710}
}
Kostenlos starten →