EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

MOSAIC: A Cognitively Motivated Multi-Agent Framework for Interpretable and Training-Free Empathetic Dialogue

Kai Liu, Hangyu Xiong, Jinyi Zhang, Min Peng

Electronics · 13.05.2026 · 15 (10) · 2078

DOI: 10.3390/electronics15102078 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Electronics, online erschienen am 13.05.2026. Verfasst von Kai Liu und 3 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 59 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Joint Laboratory on Credit Technology.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Empathetic dialogue systems built upon large language models overwhelmingly adopt a monolithic inference paradigm that processes emotion perception, causal reasoning, memory retrieval, and response planning within a single forward pass without architecturally enforced intermediate representations, forfeiting intermediate-state transparency and long-horizon personalization. Drawing on neuroscientific and cognitive–psychological evidence that human empathy is functionally dissociable, we present MOSAIC (Multi-agent Orchestration with Structured Affective memory for Interpretable empathiC dialogue), a training-free framework that operationalizes empathetic dialogue as a four-stage cognitive pipeline: affective perception, causal appraisal, episodic memory retrieval, and response synthesis. Three innovations distinguish MOSAIC from prior work: (1) a cognitively motivated modular architecture whose functionally dissociable stages enable post hoc failure attribution through logged intermediate states; (2) a hierarchical three-tier emotional memory—perceptual, semantic, and episodic—coupled with adaptive three-dimensional retrieval over emotion, situation, and coping-strategy cues; and (3) a heterogeneous model orchestration strategy coordinating open-source and API-accessible models through role-specific chain-of-thought prompts, requiring no task-specific fine-tuning. We note that the EmpatheticDialogues evaluation pre-populates the memory store with 200 training-split episodes prior to test-set interaction, a data-access asymmetry relative to single-model baselines that must be borne in mind when interpreting comparative results. Experiments on EmpatheticDialogues and ESConv show that MOSAIC achieves a 76.4% weighted F1 and an empathy score of 3.87 (on a 1–5 Likert scale) and that it improves over single-model, training-free baselines on aggregate empathy and—most prominently—on human-rated personalization (3.67 vs. 3.24 against Claude-3.5 five-shot, d=0.48). We caution that the comparison against training-free baselines is not data access-controlled (see the cold-start discussion in Methods); the personalization advantage, supported by the ablation without the Event Agent, is the result we treat as the primary practical contribution of this work.

Förderung

  • National Natural Science Foundation of China
  • Joint Laboratory on Credit Technology

Zitieren

APA

Liu, K., Xiong, H., Zhang, J., & Peng, M. (2026). MOSAIC: A Cognitively Motivated Multi-Agent Framework for Interpretable and Training-Free Empathetic Dialogue. Electronics, 15(10), 2078. https://doi.org/10.3390/electronics15102078

Vancouver

Liu K, Xiong H, Zhang J, Peng M. MOSAIC: A Cognitively Motivated Multi-Agent Framework for Interpretable and Training-Free Empathetic Dialogue. Electronics. 2026;15(10):2078. doi:10.3390/electronics15102078

BibTeX
@article{Liu2026,
  author = {Liu, Kai and Xiong, Hangyu and Zhang, Jinyi and Peng, Min},
  title = {MOSAIC: A Cognitively Motivated Multi-Agent Framework for Interpretable and Training-Free Empathetic Dialogue},
  journal = {Electronics},
  year = {2026},
  volume = {15},
  number = {10},
  pages = {2078},
  doi = {10.3390/electronics15102078}
}
Kostenlos starten →