Next Article in Journal
Helpful or Harmful? Re-Evaluating Frugality in Retrieval-Augmented Generation for Medical Question Answering
Next Article in Special Issue
Generative Artificial Intelligence and Probabilistic Trees for the Linguistic Data Summarization in Wave Energy Decision-Making
Previous Article in Journal
KhayyamNet: A Parallel Multiscale Feature Fusion Framework for Accurate Diagnosis of Multiple Sclerosis and Myelitis
Previous Article in Special Issue
Towards LLM-Driven Cybersecurity in Autonomous Vehicles: A Big Data-Empowered Framework with Emerging Technologies
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
MAKE
Volume 8
Issue 3
10.3390/make8030063
make-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Soft-Prompted Semantic Normalization for Unsupervised Analysis of the Scientific Literature

by
Ivan Malashin
*,
Dmitry Martysyuk
,
Vadim Tynchenko
*,
Andrei Gantimurov
,
Vladimir Nelyub
and
Aleksei Borodulin
Artificial Intelligence Technology Scientific and Education Center, Bauman Moscow State Technical University, 105005 Moscow, Russia
*
Authors to whom correspondence should be addressed.
Mach. Learn. Knowl. Extr. 2026, 8(3), 63; https://doi.org/10.3390/make8030063
Submission received: 29 January 2026 / Revised: 1 March 2026 / Accepted: 3 March 2026 / Published: 5 March 2026
(This article belongs to the Special Issue Using Large Language Models for Scientific Problem Solving and Engineering Design)
Browse Figures
Versions Notes

Abstract

Mapping thematic structure in large scientific corpora enables the systematic analysis of research trends and conceptual organization. This work presents an unsupervised framework that leverages large language models (LLMs) as fixed semantic inference operators guided by structured soft prompts. The framework transforms raw abstracts into normalized semantic representations that reduce stylistic variability while retaining core conceptual content. These representations are embedded into a continuous vector space, where density-based clustering identifies latent research themes without predefining the number of topics. Cluster-level interpretation is performed using LLM-based semantic decoding to generate concise, human-readable descriptions of the discovered themes. Experiments on ICML and ACL 2025 abstracts demonstrate that the method produces coherent clusters reflecting problem formulations, methodological contributions, and empirical contexts. The findings indicate that prompt-driven semantic normalization combined with geometric analysis provides a scalable and model-agnostic approach for unsupervised thematic discovery across large scholarly corpora.
Keywords: large language models; scientific literature analysis; semantic normalization; unsupervised clustering; research trend discovery large language models; scientific literature analysis; semantic normalization; unsupervised clustering; research trend discovery
Graphical Abstract

Share and Cite

MDPI and ACS Style

Malashin, I.; Martysyuk, D.; Tynchenko, V.; Gantimurov, A.; Nelyub, V.; Borodulin, A. Soft-Prompted Semantic Normalization for Unsupervised Analysis of the Scientific Literature. Mach. Learn. Knowl. Extr. 2026, 8, 63. https://doi.org/10.3390/make8030063

AMA Style

Malashin I, Martysyuk D, Tynchenko V, Gantimurov A, Nelyub V, Borodulin A. Soft-Prompted Semantic Normalization for Unsupervised Analysis of the Scientific Literature. Machine Learning and Knowledge Extraction. 2026; 8(3):63. https://doi.org/10.3390/make8030063

Chicago/Turabian Style

Malashin, Ivan, Dmitry Martysyuk, Vadim Tynchenko, Andrei Gantimurov, Vladimir Nelyub, and Aleksei Borodulin. 2026. "Soft-Prompted Semantic Normalization for Unsupervised Analysis of the Scientific Literature" Machine Learning and Knowledge Extraction 8, no. 3: 63. https://doi.org/10.3390/make8030063

APA Style

Malashin, I., Martysyuk, D., Tynchenko, V., Gantimurov, A., Nelyub, V., & Borodulin, A. (2026). Soft-Prompted Semantic Normalization for Unsupervised Analysis of the Scientific Literature. Machine Learning and Knowledge Extraction, 8(3), 63. https://doi.org/10.3390/make8030063

Article Metrics

Back to TopTop