Enhancing Language-Agnostic Systems with Slang-Aware Semantic Normalization for Robust Multilingual Understanding

  • Unique Paper ID: 207244
  • PageNo: 109-116
  • Abstract:
  • The development of language-agnostic systems has significantly advanced multilingual natural language processing by enabling shared semantic representations across languages. However, existing language-agnostic architectures primarily focus on formal linguistic structures and often struggle with informal, slang-heavy, and evolving digital communication. This limitation reduces system robustness in real-world applications such as chatbots, question-answering systems, and social media analysis. This paper proposes an enhanced language-agnostic framework incorporating a Slang-Aware Semantic Normalization (SASN) layer designed to handle millennial and Gen-Z terminology, code-mixed expressions, and informal language variations. The proposed architecture integrates universal sentence encoding, semantic role abstraction and a dynamic slang normalization module prior to semantic embedding. We introduce the concept of Informal Robustness Score (IRS) to evaluate system performance across formal, informal, and code-mixed inputs. The study demonstrates that incorporating slang-aware normalization improves semantic invariance and crosslingual transfer capability, particularly in multilingual and low-resource contexts. The proposed framework contributes toward building more inclusive and socially adaptive language-agnostic AI systems.

Copyright & License

Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

BibTeX

@article{207244,
        author = {Sayeda Mahak Musawwir and Aditya Saxena and Megha Saxena},
        title = {Enhancing Language-Agnostic Systems with Slang-Aware Semantic Normalization for Robust Multilingual Understanding},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {13},
        number = {no},
        pages = {109-116},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=207244},
        abstract = {The development of language-agnostic systems has significantly advanced multilingual natural language processing by enabling shared semantic representations across languages. However, existing language-agnostic architectures primarily focus on formal linguistic structures and often struggle with informal, slang-heavy, and evolving digital communication. This limitation reduces system robustness in real-world applications such as chatbots, question-answering systems, and social media analysis. 
This paper proposes an enhanced language-agnostic framework incorporating a Slang-Aware Semantic Normalization (SASN) layer designed to handle millennial and Gen-Z terminology, code-mixed expressions, and informal language variations. The proposed architecture integrates universal sentence encoding, semantic role abstraction and a dynamic slang normalization module prior to semantic embedding. We introduce the concept of Informal Robustness Score (IRS) to evaluate system performance across formal, informal, and code-mixed inputs. The study demonstrates that incorporating slang-aware normalization improves semantic invariance and crosslingual transfer capability, particularly in multilingual and low-resource contexts. The proposed framework contributes toward building more inclusive and socially adaptive language-agnostic AI systems.},
        keywords = {Crosslingual transfer, informal language processing, Language-agnostic systems, multilingual NLP, slang normalization, semantic role labeling},
        month = {July},
        }

Cite This Article

Musawwir, S. M., & Saxena, A., & Saxena, M. (2026). Enhancing Language-Agnostic Systems with Slang-Aware Semantic Normalization for Robust Multilingual Understanding. International Journal of Innovative Research in Technology (IJIRT), 109–116.

Related Articles