New benchmark evaluates LLMs for emotional companionship using attachment theory

A new study introduces adult attachment theory to the evaluation of large language models (LLMs) for emotional companionship. Researchers developed ECBench, a benchmark designed to assess LLMs across various relationship scenarios like emotional support and conflict resolution. The study evaluated 32 LLMs using the Experiences in Close Relationships-Revised (ECR-R) scale to measure attachment anxiety and avoidance, providing a novel framework for understanding LLM behavior in intimate contexts. AI

IMPACT This research offers a new framework for assessing LLM suitability for emotional companionship, potentially guiding development and user selection.

RANK_REASON The cluster contains an academic paper detailing a new evaluation methodology for LLMs. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CL →

• arXiv
• ECBench
• Experiences in Close Relationships-Revised (ECR-R)

• paper
• safety

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →