A new paper introduces a benchmark dataset designed to evaluate embedding models for Environmental, Social, and Governance (ESG) data. The study tested fourteen open-source and closed-source models, assessing their performance in retrieval and Retrieval-Augmented Generation (RAG) tasks specific to ESG information. Results indicated that models based on Qwen3 achieved the highest overall performance, offering practical guidance for selecting appropriate models for ESG RAG applications. AI
IMPACT Provides guidance on selecting effective embedding models for ESG data, potentially improving AI applications in sustainability and corporate accountability.
RANK_REASON Academic paper presenting a new benchmark dataset and evaluation of models for a specific domain. [lever_c_demoted from research: ic=1 ai=1.0]
Read on arXiv cs.IR (Information Retrieval) →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →