ESC-Skills：發掘與自我演進的情緒支持對話技能

摘要

現有的情感支持對話（ESC）系統主要依賴端到端的回應生成或粗略的策略監督，可解釋性有限，且難以支援系統性的技能提升。我們提出ESC-Skills，這是一個以技能為核心的架構，能夠發現並自我演化可執行的情感支持技能。我們首先將局部支持互動建模為「干預單元」（IU），捕捉尋求者狀態、支持干預行為與回應後情緒變化之間的「狀態—行動—結果」動態關係。基於從成功與失敗的ESC對話中提取的干預單元，我們建構「ESC技能庫」，其中包含干預指導、適用條件、預期結果與潛在風險等可執行的情感支持技能。為進一步提升魯棒性，我們引入多輪廓自我演化優化框架，使ESC智能體在SAGE評估下與多種模擬尋求者輪廓進行互動。所產生的互動軌跡經分析後，可識別缺失技能、不安全干預行為及特定輪廓的失敗模式，並透過基於模擬的驗證來優化技能庫。實驗結果顯示，ESC-Skills不僅提升回應層級的品質，也改善對話層級的情緒結果，同時提供更具可解釋性與可控性的支持行為。我們將於 https://github.com/aliyun/qwen-dianjin 公開程式碼、提示詞與ESC技能庫。

English

Existing emotional support conversation (ESC) systems mainly rely on end-to-end response generation or coarse strategy supervision, offering limited interpretability and little support for systematic skill improvement. We propose ESC-Skills, a skill-centric framework that discovers and self-evolves executable emotional support skills. We first model localized support interactions as Intervention Units (IUs), which capture state--action--outcome dynamics between seeker states, support interventions, and post-response emotional changes. Based on IUs extracted from both successful and failed ESC dialogues, we construct the ESC-Skills Bank, a repository of executable emotional support skills containing intervention guidance, applicability conditions, expected outcomes, and potential risks. To further improve robustness, we introduce a multi-profile self-evolutionary refinement framework in which an ESC agent interacts with diverse simulated seeker profiles under SAGE evaluation. The resulting interaction traces are analyzed to identify missing skills, unsafe interventions, and profile-specific failure patterns, which are then used to refine the Skills Bank through simulation-based verification. Experimental results demonstrate that ESC-Skills improves both response-level quality and dialogue-level emotional outcomes while providing more interpretable and controllable support behaviors. We will release the code, prompts, and ESC-Skills Bank at https://github.com/aliyun/qwen-dianjin.