Large language models for pattern recognition in text data

Kossayakova A. Ildar K. Spada L.L. Zeeshan N. Bakyt M. Khuralay M. Abdirashev O.
December 2025 Institute of Advanced Engineering and Science

IAES International Journal of Artificial Intelligence
2025 #14 Issue 6 5311 - 5332 pp.

Large language models (LLMs) are widely deployed in settings where both reliability and efficiency matter. We present a calibrated, seed-robust empirical comparison of an encoder fine-tuned model (bidirectional encoder representations from transformers (BERT)-base) and a decoder in-context model (generative pre-trained transformer (GPT)-2 small) across Stanford question answering dataset v2.0 (SQuAD v2.0) and general language understanding evaluation (GLUE)-multi-genre natural language inference (MNLI), Stanford sentiment treebank 2 (SST-2). Beyond accuracy, we assess reliability (expected calibration error with reliability diagrams and confidence–coverage analysis) and efficiency (latency, memory, throughput) under matched conditions and three fixed seeds. BERT-base yields higher accuracy and lower calibration error, while GPT-2 narrows gaps under few-shot prompting but remains more sensitive to prompt design and context length. Efficiency benchmarks show that decoder-only prompting incurs near-linear latency/memory growth with k-shot exemplars, whereas fine-tuned encoders maintain stable per-example cost. These findings offer practical guidance on when to prefer fine-tuning versus prompting and demonstrate that reliability must be evaluated alongside accuracy for risk-aware deployment.

BERT-base , Computational efficiency , Expected calibration error , GPT-2 , In context learning , Large language models , Question answering

Text of the article Перейти на текст статьи

Department of Information and Communication Technologies, M. Kozybayev North Kazakhstan University, Petropavlovsk, Kazakhstan
School of Computing, Engineering and the Built Environment, Edinburgh Napier University, Edinburgh, United Kingdom
Department of Information Security, L.N. Gumilyov Eurasian National University, Astana, Kazakhstan
Department of Space Technique and Technology, L.N. Gumilyov Eurasian National University, Astana, Kazakhstan

Department of Information and Communication Technologies
School of Computing
Department of Information Security
Department of Space Technique and Technology

10 лет помогаем публиковать статьи Международный издатель

Книга Публикация научной статьи Волощук 2026 Book Publication of a scientific article 2026