CLARIESG: An End-to-End System for ESG Analysis over Complex Tables in Corporate Reports
CLARIESG: 企業報告書の複雑な表に対するESG分析のためのエンドツーエンドシステム (AI 翻訳)
Marta Santacroce, M. Contalbo, Sara Pederzoli, Riccardo Benassi, V. Valeria, Matteo Paganelli, Francesco Guerra
🤖 gxceed AI 要約
日本語
本論文は、サステナビリティ報告書の複雑な表構造からESG情報を抽出・分析するエンドツーエンドシステム「CLARIESG」を提案。表抽出と構造化プロンプトを組み合わせ、単位正規化や文書横断推論を実現し、グリーンウォッシング検出など実用的なESG分析を支援する。
English
This paper presents CLARIESG, an end-to-end system for extracting and analyzing ESG information from complex tables in sustainability reports. It combines robust table extraction with structured prompting for unit normalization and cross-document reasoning, outperforming standard LLM methods and enabling reliable ESG analysis and greenwashing detection.
Unofficial AI-generated summary based on the public title and abstract. Not an official translation.
📝 gxceed 編集解説 — Why this matters
日本のGX文脈において
日本でもSSBJ基準に基づく有価証券報告書でのESG情報開示が進む中、複雑な表形式データの正確な抽出・分析は実務上の課題。CLARIESGはこうした開示データの信頼性向上に寄与し、投資家分析や統合報告書の検証に活用可能。
In the global GX context
With ESG disclosure frameworks like ISSB and CSRD demanding granular data, CLARIESG addresses the challenge of extracting reliable information from complex tables. Its transparent reasoning pipeline supports auditability and greenwashing detection, offering a scalable solution for global sustainability reporting.
👥 読者別の含意
🔬研究者:Provides a novel benchmarking framework and methodology for LLM-based ESG analysis over structured tables.
🏢実務担当者:Offers an automated tool to extract and verify ESG metrics from corporate reports, reducing manual effort and improving accuracy.
🏛政策担当者:Demonstrates how AI can enhance enforcement of disclosure standards and detection of misleading ESG claims.
📄 Abstract（原文）
Sustainability reports contain rich Environmental, Social and Governance (ESG) information, but their heterogeneous layouts and complex multi-table structures pose major challenges for LLMs, especially for unit normalization, cross-document reasoning, and precise numerical computation. We present C LARI ESG, an end-to-end system that couples robust table extraction with a structured prompting framework for multi-table filtering, normalization, and program-of-thought reasoning. On ESG-focused multi-table benchmarks, C LARI ESG consistently outperforms standard prompting and provides transparent, auditable reasoning, supporting more reliable ESG analysis and greenwashing detection in real-world settings.
🔗 Provenance — このレコードを発見したソース
- semanticscholar https://doi.org/10.18653/v1/2026.eacl-demo.7first seen 2026-05-15 18:39:24
🔔 こうした論文の新着を逃したくない方は キーワードアラート に登録（無料・3キーワードまで）。
gxceed は公開メタデータに基づく研究支援データセットです。要約・翻訳・解説は AI 支援で生成されています。 最終的な解釈・検証は利用者が原典資料に基づいて行うことを前提とします。