English

ObfusQAte: A Proposed Framework to Evaluate LLM Robustness on Obfuscated Factual Question Answering

Computation and Language 2026-03-05 v2 Artificial Intelligence Machine Learning

Abstract

The rapid proliferation of Large Language Models (LLMs) has significantly contributed to the development of equitable AI systems capable of factual question-answering (QA). However, no known study tests the LLMs' robustness when presented with obfuscated versions of questions. To systematically evaluate these limitations, we propose a novel technique, ObfusQAte, and leveraging the same, introduce ObfusQA, a comprehensive, first-of-its-kind framework with multi-tiered obfuscation levels designed to examine LLM capabilities across three distinct dimensions: (i) Named-Entity Indirection, (ii) Distractor Indirection, and (iii) Contextual Overload. By capturing these fine-grained distinctions in language, ObfusQA provides a comprehensive benchmark for evaluating LLM robustness and adaptability. Our study observes that LLMs exhibit a tendency to fail or generate hallucinated responses when confronted with these increasingly nuanced variations. To foster research in this direction, we make ObfusQAte publicly available.

Keywords

Cite

@article{arxiv.2508.07321,
  title  = {ObfusQAte: A Proposed Framework to Evaluate LLM Robustness on Obfuscated Factual Question Answering},
  author = {Shubhra Ghosh and Abhilekh Borah and Aditya Kumar Guru and Kripabandhu Ghosh},
  journal= {arXiv preprint arXiv:2508.07321},
  year   = {2026}
}

Comments

LREC 2026