EmeraldMind: A Knowledge Graph-Augmented Framework for Greenwashing Detection

TL;DR

EmeraldMind is a novel framework that combines a domain-specific knowledge graph with retrieval-augmented generation to detect greenwashing claims, providing transparent, evidence-backed verdicts and improving accuracy and coverage over generic models.

Contribution

It introduces EmeraldMind, a fact-centric, knowledge graph-augmented system for greenwashing detection that does not require fine-tuning of large language models.

Findings

Achieves competitive accuracy in greenwashing detection

Provides greater coverage and explanation quality

Operates without fine-tuning or retraining of models

Abstract

As AI and web agents become pervasive in decision-making, it is critical to design intelligent systems that not only support sustainability efforts but also guard against misinformation. Greenwashing, i.e., misleading corporate sustainability claims, poses a major challenge to environmental progress. To address this challenge, we introduce EmeraldMind, a fact-centric framework integrating a domain-specific knowledge graph with retrieval-augmented generation to automate greenwashing detection. EmeraldMind builds the EmeraldGraph from diverse corporate ESG (environmental, social, and governance) reports, surfacing verifiable evidence, often missing in generic knowledge bases, and supporting large language models in claim assessment. The framework delivers justification-centric classifications, presenting transparent, evidence-backed verdicts and abstaining responsibly when claims cannot be verified. Experiments on a new greenwashing claims dataset demonstrate that EmeraldMind achieves competitive accuracy, greater coverage, and superior explanation quality compared to generic LLMs, without the need for fine-tuning or retraining.