I Understand How You Feel: Enhancing Deeper Emotional Support Through Multilingual Emotional Validation in Dialogue System
For researchers in multilingual dialogue systems, this work provides resources and a model for emotional validation, though the gains are incremental.
The paper introduces M-EDESConv, a 120k English-Japanese multilingual corpus for emotional validation in dialogue systems, and proposes MEGUMI, a model that fuses XLM-RoBERTa semantics with language-specific emotion encoders. MEGUMI achieves superior performance on both M-EDESConv and M-TESC datasets, while benchmarks of GPT-4.1 Nano and Llama-3.1 8B show current LLMs generate contextually similar responses but lack emotional understanding.
Emotional validation - explicitly acknowledging that a user's feelings make sense - has proven therapeutic value but has received little computational attention. Emotional validation in dialogue systems can be decomposed into (i) validating response identification, (ii) validation timing detection, and (iii) validating response generation. To support research on all three subtasks, we release M-EDESConv, a 120k English-Japanese multilingual corpus created through hybrid manual and automatic annotation, and M-TESC, a multilingual spoken-dialogue test set. For timing detection, we propose MEGUMI, a Multilingual Emotion-aware Gated Unit for Mutual Integration, that fuses frozen XLM-RoBERTa semantics with language-specific emotion encoders via cross-modal attention and gated fusion. MEGUMI shows superior performance on both the M-EDESConv and M-TESC datasets, both objectively and subjectively. Finally, our EmoValidBench benchmarks of GPT-4.1 Nano and Llama-3.1 8B indicate that current LLMs generate contextually similar and diverse validating responses, but emotional understanding remains a major area for improvement. Project page: https://github.com/zihaurpang/Multilingual-Emotional-Validation