Bias Beneath the Tone: Empirical Characterisation of Tone Bias in LLM-Driven UX Systems
Authors:
- Heet BodaraHBHeet Bodara
- Masum MushfiqMMMasum Mushfiq
- Isma Farah SiddiquiISIsma Farah Siddiqui
Preprint
Abstract
Large Language Models are increasingly used in conversational systems such as digital personal assistants, shaping how people interact with technology through language. While their responses often sound fluent and natural, they can also carry subtle tone biases such as sounding overly polite, cheerful, or cautious even when neutrality is expected. These tendencies can influence how users perceive trust, empathy, and fairness in dialogue. In this study, we explore tone bias as a hidden behavioral trait of large language models. The novelty of this research lies in the integration of controllable large language model based dialogue synthesis with tone classification models, enabling robust and ethical emotion recognition in personal assistant interactions. We created two synthetic dialogue datasets, one generated from neutral prompts and another explicitly guided to produce positive or negative tones. Surprisingly, even the neutral set showed consistent tonal skew, suggesting that bias may stem from the model's underlying conversational style. Using weak supervision through a pretrained DistilBERT model, we labeled tones and trained several classifiers to detect these patterns. Ensemble models achieved macro F1 scores up to 0.92, showing that tone bias is systematic, measurable, and relevant to designing fair and trustworthy conversational AI.Keywords
All Related Versions
This publication has 15 references indexed in Scilit:
- AI–AI bias: Large language models favor communications generated by large language modelsProceedings of the National Academy of Sciences, 2025
- Enhanced Emotion-Aware Conversational Agent: Analyzing User Behavioral Status for Tailored Reponses in Chatbot InteractionsIEEE Access, 2025
- A comprehensive survey of text classification techniques and their research applications: Observational and experimental insightsComputer Science Review, 2024
- A survey of dialogic emotion analysis: Developments, approaches and perspectivesPattern Recognition, 2024
- CFN-ESA: A Cross-Modal Fusion Network With Emotion-Shift Awareness for Dialogue Emotion RecognitionIEEE Transactions on Affective Computing, 2024
- Multimodal Knowledge-enhanced Interactive Network with Mixed Contrastive Learning for Emotion Recognition in ConversationNeurocomputing, 2024
- Emotion Recognition in Conversation Based on a Dynamic Complementary Graph Convolutional NetworkIEEE Transactions on Affective Computing, 2024
- Dynamic interactive multiview memory network for emotion recognition in conversationInformation Fusion, 2022
- BiERU: Bidirectional emotional recurrent unit for conversational sentiment analysisNeurocomputing, 2022
- The optimally designed dynamic memory networks for targeted sentiment classificationNeurocomputing, 2018