Authors : Kapil Kumar Goyal

Volume/Issue : Volume 10 - 2025, Issue 5 - May

Google Scholar : https://tinyurl.com/3bksasar

DOI : https://doi.org/10.38124/ijisrt/25may1438

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Abstract : Large Language Models (LLMs) have shown significant promise in enhancing enterprise productivity across domains like customer service, document summarization, and decision support. However, their performance is highly dependent on the structure and phrasing of input prompts. This paper proposes a novel framework called Prompt Elasticity, which introduces adaptive input shaping mechanisms based on contextual factors such as user intent, domain specificity, and prior interaction history. We detail the architectural components of this framework, present a prototype implementation in a customer support environment, and demonstrate improvements in both reliability and relevance of LLM outputs. Our results show a measurable uplift in response quality and user satisfaction. The proposed framework offers a lightweight, scalable addition to enterprise LLM workflows that enhances both performance and interpretability.

Keywords : Prompt Engineering, LLM, Input Shaping, Enterprise AI, Context-Aware NLP, Adaptive Systems.

References :

1. T. B. Brown, B. Mann, N. Ryder, M. Subbiah, J. D. Kaplan, et al., “Language Models are Few-Shot Learners,” in Advances in Neural Information Processing Systems (NeurIPS), vol. 33, pp. 1877–1901, 2020
2. L. Reynolds and K. McDonell, “Prompt Programming for Large Language Models: Beyond the Few-Shot Paradigm,” arXiv preprint arXiv:2102.07350, 2021. [Online]. Available: https://arxiv.org/abs/2102.07350
3. H. Yang, D. Lin, and M. Tan, “Structured Prompting: Bridging the Gap Between Natural and Symbolic Reasoning in LLMs,” arXiv preprint arXiv:2505.13406, May 2025. [Online]. Available: https://arxiv.org/abs/2505.13406
4. C. Raffel, N. Shazeer, A. Roberts, K. Lee, S. Narang, et al., “Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer,” J. Mach. Learn. Res., vol. 21, no. 140, pp. 1–67, 2020.
5. H. Yang, M. Gupta, and S. Desai, “Context-aware prompting improves clinical information extraction from patient-provider messages,” JAMIA Open, vol. 7, no. 3, ooae080, 2024. [Online]. Available: https://doi.org/10.1093/jamiaopen/ooae080
6. X. Zhou, J. Zhang, C. Li, Y. Lu, and M. Li, “PromptBench: Benchmarking Prompt Engineering for Large Language Models,” arXiv preprint arXiv:2406.05673, 2024. [Online]. Available: https://arxiv.org/abs/2406.05673
7. M. Liu, B. Liang, M. Zhang, Y. Yang, and T.-Y. Liu, “A systematic survey of prompt engineering in large language models: Techniques and applications,” arXiv preprint arXiv:2410.23405, Oct. 2024. [Online]. Available: https://arxiv.org/abs/2410.23405
8. P. Sahoo, A. K. Singh, S. Saha, V. Jain, S. Mondal, and A. Chadha, “A systematic survey of prompt engineering in large language models: Techniques and applications,” arXiv preprint arXiv:2402.07927, Feb. 2024. [Online]. Available: https://arxiv.org/abs/2402.07927
9. Y. Zhang, J. Wang, X. Li, and M. Wang, “P-Eval: A Comprehensive Evaluation Framework for Prompt Engineering in LLMs,” arXiv preprint arXiv:2505.13416, 2025. [Online]. Available: https://arxiv.org/abs/2505.134