Detecting Emotional Dynamic Trajectories: An Evaluation Framework for Emotional Support in Language Models

TL;DR

This paper introduces a trajectory-based evaluation framework for assessing large language models' ability to provide long-term emotional support, emphasizing dynamic emotional state tracking over static, short-term assessments.

Contribution

It presents a novel benchmark with metrics for evaluating LLMs on their capacity to sustain and stabilize user emotions over time using realistic dialogue scenarios.

Findings

Significant differences in emotional support capabilities among evaluated LLMs.

The framework effectively captures long-term emotional dynamics and model performance.

Guidelines for improving emotionally supportive responses in LLMs.

Abstract

Emotional support is a core capability in human-AI interaction, with applications including psychological counseling, role play, and companionship. However, existing evaluations of large language models (LLMs) often rely on short, static dialogues and fail to capture the dynamic and long-term nature of emotional support. To overcome this limitation, we shift from snapshot-based evaluation to trajectory-based assessment, adopting a user-centered perspective that evaluates models based on their ability to improve and stabilize user emotional states over time. Our framework constructs a large-scale benchmark consisting of 328 emotional contexts and 1,152 disturbance events, simulating realistic emotional shifts under evolving dialogue scenarios. To encourage psychologically grounded responses, we constrain model outputs using validated emotion regulation strategies such as situation selection and cognitive reappraisal. User emotional trajectories are modeled as a first-order Markov process, and we apply causally-adjusted emotion estimation to obtain unbiased emotional state tracking. Based on this framework, we introduce three trajectory-level metrics: Baseline Emotional Level (BEL), Emotional Trajectory Volatility (ETV), and Emotional Centroid Position (ECP). These metrics collectively capture user emotional dynamics over time and support comprehensive evaluation of long-term emotional support performance of LLMs. Extensive evaluations across a diverse set of LLMs reveal significant disparities in emotional support capabilities and provide actionable insights for model development.