ESG-Bench: Benchmarking Long-Context ESG Reports for Hallucination Mitigation

TL;DR

This paper introduces ESG-Bench, a benchmark dataset for evaluating and improving large language models' ability to interpret complex ESG reports accurately and mitigate hallucinations, enhancing trustworthy automation in ESG analysis.

Contribution

The paper presents ESG-Bench, a new dataset with human-annotated QA pairs and verifiability labels, along with CoT prompting strategies and fine-tuning methods to reduce hallucinations in LLMs for ESG report understanding.

Findings

CoT prompting significantly reduces hallucinations in LLMs.

Fine-tuning on ESG-Bench improves factual accuracy of model outputs.

Method transfers improvements to other QA benchmarks.

Abstract

As corporate responsibility increasingly incorporates environmental, social, and governance (ESG) criteria, ESG reporting is becoming a legal requirement in many regions and a key channel for documenting sustainability practices and assessing firms' long-term and ethical performance. However, the length and complexity of ESG disclosures make them difficult to interpret and automate the analysis reliably. To support scalable and trustworthy analysis, this paper introduces ESG-Bench, a benchmark dataset for ESG report understanding and hallucination mitigation in large language models (LLMs). ESG-Bench contains human-annotated question-answer (QA) pairs grounded in real-world ESG report contexts, with fine-grained labels indicating whether model outputs are factually supported or hallucinated. Framing ESG report analysis as a QA task with verifiability constraints enables systematic evaluation of LLMs' ability to extract and reason over ESG content and provides a new use case: mitigating hallucinations in socially sensitive, compliance-critical settings. We design task-specific Chain-of-Thought (CoT) prompting strategies and fine-tune multiple state-of-the-art LLMs on ESG-Bench using CoT-annotated rationales. Our experiments show that these CoT-based methods substantially outperform standard prompting and direct fine-tuning in reducing hallucinations, and that the gains transfer to existing QA benchmarks beyond the ESG domain.