Problem
Research questions and friction points this paper is trying to address.
Emotional Understanding
Conversation Relevance
Predictive Capabilities
Sibyl: Empowering Empathetic Dialogue Generation in Large Language Models via Sensible and Visionary Commonsense Inference
🤖 AI Summary
Existing large language models struggle with dynamic empathy in multi-turn dialogues, particularly lacking the capacity to anticipate interlocutor emotions and infer underlying needs in subsequent utterances. To address this, we propose the Sensible and Visionary Commonsense (SVC) reasoning framework—a novel approach that explicitly models commonsense knowledge conditioned on the *next* dialogue turn. SVC jointly leverages context-aware representation learning and causal inference to dynamically extract and selectively inject dialogue-level commonsense, enabling proactive empathic response generation. Unlike conventional static commonsense injection methods, SVC enables turn-level adaptivity and forward-looking reasoning. Evaluated on multiple empathetic dialogue benchmarks, SVC achieves state-of-the-art performance; human evaluations demonstrate an average 23.6% improvement in empathy, supportiveness, and coherence scores.
📝 Abstract
Recently, there has been a heightened interest in building chatbots based on Large Language Models (LLMs) to emulate human-like qualities in multi-turn conversations. Despite having access to commonsense knowledge to better understand the psychological aspects and causality of dialogue context, even these powerful LLMs struggle to achieve the goals of empathy and emotional support. Current commonsense knowledge derived from dialogue contexts is inherently limited and often fails to adequately anticipate the future course of a dialogue. This lack of foresight can mislead LLMs and hinder their ability to provide effective support. In response to this challenge, we present an innovative framework named Sensible and Visionary Commonsense Knowledge (Sibyl). Designed to concentrate on the immediately succeeding dialogue, this paradigm equips LLMs with the capability to uncover the implicit requirements of the conversation, aiming to elicit more empathetic responses. Experimental results demonstrate that incorporating our paradigm for acquiring commonsense knowledge into LLMs comprehensively enhances the quality of their responses.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Sibyl Method
Large Language Models
Enhanced Emotional Understanding
🔎 Similar Papers
No similar papers found.
