MICA combines incremental per-turn distance rewards and Monte Carlo returns from a shared potential function over user support states to create a mixed advantage signal that enables stable multi-turn RL optimization for emotional support dialogues.
Liu, Jinfeng Zhou, Alvionna S
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
Authors build an emotional intensity dataset and fine-tune generative LLMs to predict continuous 0-100 scores, claiming outperformance over classification baselines plus generalization to sentiment and arousal.
A holistic survey of affective computing for intelligent agents covering emotion understanding via multimodal data, affective cognition, emotional expression synthesis, key challenges, and future directions emphasizing generative technologies.