Personalized Emotional Intelligence in Generative AI through Symbolic Affective Reasoning

๐Ÿ“… 2026-07-12
๐Ÿ“ˆ Citations: 0
โœจ Influential: 0
๐Ÿ“„ PDF
๐Ÿค– AI Summary
Current generative AI systems struggle to understand, predict, and proactively guide users toward desired emotional states. This work proposes EROS, a novel framework that integrates symbolic affective reasoning with deep generative models for the first time. EROS leverages a scalable memory bank to enable instant personalization without fine-tuning and employs attention mechanisms to localize emotion-relevant regions in images, thereby generating visually coherent content that preserves semantic fidelity while aligning with target emotions. The model learns generalizable affective rules from large-scale imageโ€“emotion datasets and constructs interpretable emotional profiles tailored to individual preferences. Psychophysical experiments demonstrate that EROS significantly outperforms state-of-the-art multimodal large models in both eliciting intended emotions and adapting to individual differences.
๐Ÿ“ Abstract
Emotional intelligence enables humans to recognize emotions, infer their causes, reason about interventions, and modify their environment to achieve desired affective states. Despite recent advances in artificial intelligence (AI), current models remain largely limited to generating realistic content or performing semantic reasoning, with little capacity for understanding, predicting, and personalizing human emotional responses. Here we introduce Emotion-augmented geneRatiOn System (EROS), a hybrid AI framework that integrates symbolic reasoning with deep learning to enable personalized emotion augmentation through visual content. Leveraging large-scale image-emotion datasets, EROS discovers generalizable affective rules, identifies emotion-relevant image regions, and predicts context-aware visual modifications that preserve scene semantics while steering emotional responses toward desired targets. To account for individual variability, EROS incorporates an expandable memory bank that supports inference-time personalization without model fine-tuning, yielding interpretable emotional profiles and rapid adaptation to new users. Across extensive human psychophysics experiments, EROS elicits target emotional responses more effectively than state-of-the-art large multimodal models while adapting to individual affective preferences. Beyond affective computing, EROS provides a foundation for AI systems that can understand, reason about, and augment human cognitive states, with potential applications in mental health, adaptive media, education, and human-computer interaction.
Problem

Research questions and friction points this paper is trying to address.

emotional intelligence
personalization
affective computing
emotion prediction
human-AI interaction
Innovation

Methods, ideas, or system contributions that make the work stand out.

symbolic affective reasoning
personalized emotional intelligence
hybrid AI framework
emotion-aware image generation
inference-time personalization
๐Ÿ”Ž Similar Papers
No similar papers found.