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

Unofficial AI-generated summary based on the public title and abstract. Not an official translation.

🔬研究者:Researchers can use ESG-Bench to evaluate and improve LLMs for ESG-specific tasks.

🏢実務担当者:Corporate sustainability teams can leverage this benchmark to assess AI tools for ESG report analysis.

🏛政策担当者:Policymakers can consider this benchmark when setting standards for AI-assisted disclosure verification.

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.

• semanticscholar https://doi.org/10.1609/aaai.v40i46.41281first seen 2026-05-15 18:29:18