Skip to content
Open access

BiasScope: Inference-Time Bias Detection and Mitigation for Fair Natural Language Processing (NLP) Using Retrieval-Augmented Generation and Prompt Engineering

Sep 2026 · International Journal of Intelligent Systems and Data Science · 2 citations · 23 references

TL;DR

BiasScope is a lightweight framework that uses prompt engineering and RAG to detect and mitigate model biases at inference time without changing weights, and provides an effective means of achieving fairness and accountability in production Natural Language Processing (NLP).

Abstract

The appearance of social stereotypes in Large Language Models (LLMs) is a crucial concern in research on fairness in the context of Natural Language Processing (NLP). This study focuses on gender bias, occupational stereotypes, gender comparisons, and trait attributions. Numerous mitigation techniques depend on fine-tuning and post-processing, which are typically expensive and difficult to audit. BiasScope is a lightweight framework that uses prompt engineering and RAG to detect and mitigate model biases at inference time without changing weights. We generate responses through the Gemini 2.5 Flash API in our implementation. The system (1) provides a baseline reply, followed by (2) scoring by a lexicon- and pattern-based detector with transparent functioning. When bias is detected, (3) it retrieves fairness guidance, followed by (4) generating a mitigated response from an augmented prompt. The mean composite bias score on the 26-prompt evaluation set is reduced by 72.1% after mitigation. This composite score combines four detector signals: toxicity, counterfactual sensitivity, stereotype co-occurrence, and sentiment and sentiment is treated as one signal among four, not as proof of bias. No formal user study of response quality was conducted; qualitative inspection of the outputs is reported in Section 4. BiasScope is designed to support multiple LLM backends and is fully traceable from detection to mitigation. It provides an effective means of achieving fairness and accountability in production Natural Language Processing (NLP).

Read PDF

Similar papers

Open access Aug 2026

Bias Detection and Optimization of Recruitment Texts Based on BERT

The proposed framework provides a reliable and interpretable approach for automated fairness correction of recruitment texts and offers practical guidance for intelligent text processing and trustworthy information management in engineering applications.

L.-P. Feng, N. Feng, Q.-J. Hui · 0 citations
Open access Oct 2026

CoR-Hate: counterfactual retrieval and fairness-regularized hate-speech detection

The growing presence of hate speech on social media, targeting groups based on race, religion, gender, and sexual orientation, raises critical concerns regarding fairness and reliability in automated moderation systems. This study introduces CoR-Hate, a fairness-regularized framework that integrates Sentence-Transf...

Ehtesham Hashmi, J. Mccrae, Sule Yildirim Yayilgan et al. · 0 citations
#artificial intelligence Conference Open access Sep 2026

Linguistic Triggers of Gender and Racial Bias in Open-Weight LLMs Applied to Recruitment

Open-weight large language models are rapidly entering hiring pipelines, yet their discriminatory failure modes—and the regulatory exposure these create under the EU AI Act high-risk classification (Annex III) and U.S. EEOC adverse-impact analysis—remain poorly understood. We present the first systematic, multi-model a...

Kosuke Kitahara, Nobuhiro Yamaguchi · 1 citation · ⚡1
Aug 2026

How prompts manipulate models: measuring cognitive bias vulnerability in commercial LLMs across languages

It is observed that bias vulnerability increases in Hindi and Bengali compared to English, particularly under manipulative prompts, which highlights the importance of multilingual bias evaluation and provides practical guidance for selecting commercial language models in bias-sensitive applications.

Koushik Deb, Imon Mukherjee, Debarshi Kumar Sanyal · 0 citations
Sep 2026

Adaptive NLI-Driven Claim Verification with Statistical Decision Modeling for Low-Latency Hallucination Reduction in Large Language Models.

A lightweight two-step claim verification framework that decomposes LLM responses into atomic factual claims and independently verifies each extracted claim against a separately generated reference produced through an isolated factual recall prompt, showing consistent performance across the evaluated benchmarks without...

Subasish Mohapatra, Biswajeet Dash, Subhadarshini Mohanty et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.