TEA-Bench: A Systematic Benchmarking of Tool-enhanced Emotional Support Dialogue Agent
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914541545193472 |
|---|---|
| author | Sui, Xingyu Zhao, Yanyan Hu, Yulin Guo, Jiahe Zhao, Weixiang Qin, Bing |
| author_facet | Sui, Xingyu Zhao, Yanyan Hu, Yulin Guo, Jiahe Zhao, Weixiang Qin, Bing |
| contents | Emotional Support Conversation requires not only affective expression but also grounded instrumental support to provide trustworthy guidance. However, existing ESC systems and benchmarks largely focus on affective support in text-only settings, overlooking how external tools can enable factual grounding and reduce hallucination in multi-turn emotional support. We introduce TEA-Bench, the first interactive benchmark for evaluating tool-augmented agents in ESC, featuring realistic emotional scenarios, an MCP-style tool environment, and process-level metrics that jointly assess the quality and factual grounding of emotional support. Experiments on nine LLMs show that tool augmentation generally improves emotional support quality and reduces hallucination, but the gains are strongly capacity-dependent: stronger models use tools more selectively and effectively, while weaker models benefit only marginally. We further release TEA-Dialog, a dataset of tool-enhanced ESC dialogues, and find that supervised fine-tuning improves in-distribution support but generalizes poorly. Our results underscore the importance of tool use in building reliable emotional support agents. Our code and data can be found in https://github.com/XingYuSSS/TEA-Bench. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_18700 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | TEA-Bench: A Systematic Benchmarking of Tool-enhanced Emotional Support Dialogue Agent Sui, Xingyu Zhao, Yanyan Hu, Yulin Guo, Jiahe Zhao, Weixiang Qin, Bing Artificial Intelligence Emotional Support Conversation requires not only affective expression but also grounded instrumental support to provide trustworthy guidance. However, existing ESC systems and benchmarks largely focus on affective support in text-only settings, overlooking how external tools can enable factual grounding and reduce hallucination in multi-turn emotional support. We introduce TEA-Bench, the first interactive benchmark for evaluating tool-augmented agents in ESC, featuring realistic emotional scenarios, an MCP-style tool environment, and process-level metrics that jointly assess the quality and factual grounding of emotional support. Experiments on nine LLMs show that tool augmentation generally improves emotional support quality and reduces hallucination, but the gains are strongly capacity-dependent: stronger models use tools more selectively and effectively, while weaker models benefit only marginally. We further release TEA-Dialog, a dataset of tool-enhanced ESC dialogues, and find that supervised fine-tuning improves in-distribution support but generalizes poorly. Our results underscore the importance of tool use in building reliable emotional support agents. Our code and data can be found in https://github.com/XingYuSSS/TEA-Bench. |
| title | TEA-Bench: A Systematic Benchmarking of Tool-enhanced Emotional Support Dialogue Agent |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2601.18700 |