EmoTrace, a multi-turn dialogue corpus generation framework centered on modeling seekers' emotional trajectories, is proposed, which outperforms existing approaches in terms of emotional richness and empathy quality.
Abstract
Using large language models (LLMs) to assist psychological counseling is an important task in the field of natural language processing. The construction of high-quality psychological support dialogue corpora serves as a critical foundation for training counseling-oriented conversational models. However, existing data generation approaches generally suffer from several limitations, including emotionally stable seekers, limited variation in emotional dynamics, and a high degree of compliance with counselors'guidance. These issues result in LLM that lack the capability to effectively respond to emotionally unstable scenarios. In addition, counselor responses are typically driven by problem-solving objectives, thereby overlooking the role of emotion-focused interaction, which are essential in psychological counseling. To address these gaps, we propose EmoTrace, a multi-turn dialogue corpus generation framework centered on modeling seekers'emotional trajectories. we construct seekers'cognitive profile and introduce a seeker module with emotional schemas and an associated activation mechanism, a counselor module, and an emotional trajectory control module, thereby enhancing the layering of the seeker's emotional expression and the counselor's targeted empathic expression. Experimental results demonstrate that the proposed method outperforms existing approaches in terms of emotional richness and empathy quality.
Large Language Models (LLMs) have substantially advanced persona-based dialogue agents for emotion-sensitive role simulation in healthcare, education, counseling, customer service, and interactive storytelling. However, two related lines of work leave a key gap. Persona-based dialogue systems often encode emotions as static traits or surface-level stylistic cues, and affective dialogue research has largely focused on empathetic response generation toward users rather than modeling the agent persona's own evolving emotional state. As a result, trigger-driven emotional evolution within a character remains underexplored. To address this limitation, we draw inspiration from the Component Process Model (CPM), a psychological theory that views emotion as a dynamic process shaped by the appraisal of external events. We propose CPM-MultiAgent, a CPM-grounded emotion evolution multi-agent framework for supporting emotional changes in persona-based dialogue. Instead of treating a character's emotion as a fixed attribute, CPM-MultiAgent represents it as a latent state that is continuously reshaped by dialogue triggers. Through affective trigger extraction, CPM-based collaborative appraisal, and emotion state updating, the framework enables more emotionally consistent role simulation in multi-turn interactions.Experiments with baseline comparisons, ablation studies, human evaluation, and case analyses demonstrate that CPM-MultiAgent effectively models dynamic emotional evolution in emotionally sensitive role-simulation settings.
Jingyao Cai, Shuaijun Liu, Abdul Rehman et al.· 0 citations
Emotional dialogue research includes two influential strategy traditions. Empathetic dialogue prioritizes understanding a speaker's emotional experience. Emotional support conversation selects and sequences support for the seeker's current needs. Sustained use introduces a further goal. Effective support should sustain users'capacities for emotion regulation, coping, self-endorsed decisions, and social connection across the interaction lifecycle. We propose capability-sustaining emotional dialogue (CSED) as a longitudinal research paradigm that aligns supportive strategy with this goal and organizes data, models, system design, evaluation, and governance around repeated use, non-use, transition, and termination. A targeted literature-and-corpus audit motivates this position. In a PRISMA-ScR-guided sample, 95% of 60 system-building papers pursue relief-oriented goals. None evaluates capability or longitudinal outcomes, and only 1 considers dependency, autonomy, or termination risk. In 300 ESConv supporter turns, capability-relevant functions appear in 43.0%, while generic suggestions account for 22.0%, compared with 4.0% reappraisal, 6.7% self-efficacy support, and 0.3% boundary behavior. We release a protocol for extending the audit to model behavior. An illustrative process model connects latent user capability to six design commitments, four evaluation timescales, and lifecycle constraints. The resulting agenda makes CSED testable across data, policy design, training, evaluation, and governance.
Ming Wang, Jiaqi Wu Young, Wenfang Wu et al.· 0 citations
Existing dialogue resources for minors mainly focus on general psychological support, adolescent positive mental health promotion, or broad emotional companionship, and therefore do not adequately support the task of everyday emotion coaching for Chinese children. To address this gap, we construct ChildEC, a developmentally stratified multi-turn dialogue corpus for everyday emotion coaching, targeting Chinese children aged 8--12. The corpus contains 1,709 multi-turn dialogues spanning two developmental stages and five core themes. To support corpus construction and evaluation, we further propose the Theory-Grounded Emotion Coaching Framework (TGEC). Within this framework, TGEC-Synth operationalizes the five-step emotion coaching model and Socratic dialogue into a multi-stage dialogue synthesis pipeline, while TGEC-Eval assesses the quality of child emotion coaching dialogues along six key dimensions. Experimental results show that models fine-tuned on ChildEC consistently outperform their corresponding base models under both automatic metrics and LLM-as-a-Judge evaluation, demonstrating the downstream training value of the corpus for child emotion coaching. Overall, ChildEC provides a specialized data resource and an evaluation reference for developmentally sensitive and theory-consistent research on Chinese child emotion coaching. Our dataset will be publicly released upon acceptance of the paper.
Yuelin Ding· Poster Volume 0007 The 2026...· 0 citations
We developed a large language model designed to explore student mental well-being support in conversational settings, aimed at providing accessible, empathetic, and accurate responses to students facing challenges such anxiety, stress, loneliness, and academic pressure. Many students face barriers to seeking traditional counseling, such as stigma, scheduling constraints or discomfort with face-to-face interactions. The system addresses these challenges by offering a potential accessible conversational support channel through natural, conversational interactions with a large language model. Our approach focused on two strategies. Firstly, we enhanced the model's communication style to reflect counseling best practices such as empathy, active listening, and emotional validation. The second strategy is to enhance the model's understanding of mental health scenarios using realworld text sources and instructions. The model was trained on diverse, anonymized datasets from real counseling transcripts, emotional support dialogue corpora, and peer-support forums. We integrated prompt engineering, fine-tuning, and an iterative self-reflection loop to identify potentially unsupported or hallucination-prone responses, with the goal to improve factuality and safety in generated responses. We find that fine-tuning on student-centered data consistently outperforms both baseline and mixed-data approaches, emphasizing the importance of domainspecific adaptation. The model shows potential for confidential support, suggesting possible use as an early stage aid for coping strategies, and connects students to campus resources, reducing barriers to help seeking and supporting academic performance.
Sarthak Musmade, Lu Liu· International Conference on...· 0 citations
Recent advances in artificial intelligence have enabled the development of more natural and effective human-machine interactions. In this study, we propose an enhanced BART-based dialogue generation framework that integrates emotion labels and keyword annotations to support emotionally aware end-to-end conversations. The proposed system combines an independent RoBERTa-based emotion classification module with a terminology-aware annotation mechanism, in which emotional labels and key semantic terms are explicitly injected into the dialogue context. The framework is evaluated on a merged dataset constructed from two benchmark dialogue corpora, Empathetic Dialogues and DailyDialog, which have been relabeled into nine unified emotion categories. Automatic evaluation is conducted using BLEU, ROUGE, and perplexity metrics. Experimental results demonstrate that the proposed approach consistently outperforms baseline models, including GPT-2, DialoGPT, T5, UniLM, and the original BART model. In particular, the incorporation of keyword annotations significantly improves response relevance and informational completeness, while emotion labels further enhance emotional alignment. These findings suggest that explicitly modeling emotional cues and semantic focus can effectively improve the quality of emotional dialogue generation, offering a promising direction for the development of emotionally intelligent conversational agents.
Emotional Support (ES) systems have long optimized a single objective: alleviating the user's emotional distress in the moment. We argue that a complementary need, helping users see themselves more clearly, defines a distinct paradigm we call Personality Support (PS). PS is not counseling or clinical intervention: it targets cognitive clarity and self-articulation, not symptom relief or diagnosis. We instantiate this paradigm in three layers. First, we present DSD, a Chinese self-discovery PS Dataset of 8,590 samples collected through real longitudinal interaction across five minimal units, Coach, Warm, Tsukkomi, Real, and Gonzo. Second, we build DeepSupport, a multi-persona PS system trained with OrthoTune, a PS-tailored framework with style-specific adapters and a style-consistency regularizer. Third, we unify the five DeepSupport personas into Ekova, a persistent personality-support agent with a unified cross-session memory layer, supporting both adaptive routing and user-customized persona selection. Experiments show that OrthoTune-trained models outperform all baselines with an average relative gain of 16.3% across all metrics over the strongest prompt-based baseline. Code is available at https://github.com/Yukyin/Ekova.