EmoStance：用 emoji 弱监督实现共情回复生成中的回复侧情感取向控制

View PDF HTML (experimental)

Abstract:Empathetic response generation requires models to decide not only what to say, but also how to respond to the previous speaker's affective situation. We formulate this as response-side affective-orientation control and use multi-annotator emoji distributions as weak affective--attitudinal evidence, rather than as output symbols or gold labels, to induce a latent control space that operationally approximates listener stance. We construct EmojiDialogue, an utterance-level extension of EmpatheticDialogues with emoji votes and confidence scores, and propose EmoStance, which models source-side affective expression, predicts a soft response-side orientation from dialogue context and speaker roles, and steers a frozen instruction-tuned LLM through continuous prefix embeddings. In blind pairwise evaluation with 20 annotators and 800 judgments, EmoStance achieves a 62.2% decisive win rate, with the clearest gains in contextual specificity and perceived responsiveness, while remaining complementary to external-knowledge methods. Code, annotation metadata, and reconstruction scripts are available in our GitHub repository: this https URL.

Comments:	Accepted to the Main Conference of EMNLP 2026(Oral Presentation); manuscript unchanged from v1
Subjects:	Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
Cite as:	arXiv:2609.02133 [cs.AI]
	(or arXiv:2609.02133v2 [cs.AI] for this version)
	https://doi.org/10.48550/arXiv.2609.02133 arXiv-issued DOI via DataCite

Submission history

From: Ziyuan Jin [view email]
[v1] Wed, 2 Sep 2026 05:43:14 UTC (1,869 KB)
[v2] Mon, 12 Oct 2026 14:58:06 UTC (1,869 KB)