GX & Decarbonization Research

Francisco Javier Rodríguez-Ruiz, Ana María García-Berbaneu, Alexsander Luiz Telpisow-Scheid

This study proposes an automated methodology using NLP to extract, classify, and assess ESG content from corporate sustainability reports. It uses a taxonomy aligned with European reporting standards for listed SMEs, enabling scalable and r…

Amir Amel-Zadeh, Tami Dinh, Andreas Seebeck +1

This paper uses deep learning to analyze visuals and text in 3,923 European sustainability reports (2013-2021), documenting a functional separation between graphics and photographs. Firms with stronger ESG performance use more graphics but …

Mrs. Awantika Deshpande, Mr. Kiran More

This paper proposes an AI-driven education system to enhance ESG reporting knowledge and consistency. It integrates NLP, machine learning, and knowledge graphs for framework analysis, adaptive learning, and real-time reporting assistance. E…

Iulia Florentina Voicila Voicila, Elena UrquiaGrande

Using 87,225 private firms from Spain and the UK, this study shows ESG indicators improve financial distress prediction only when disclosure quality is high (UK). In Spain, fragmented ESG data yields no improvement. Machine learning models,…

Stefan Vieweg, Christoph Klein

This review synthesizes 2022-2025 evidence on AI in green finance and ESG, proposing the RESP-ESG framework for responsible adoption. It maps eight archetypes, showing NLP and ML enhance ESG signal extraction and climate analytics, while hi…

Ambika Zutshi, Diane Zandee, Andrew Creed

This paper positions AI as a critical enabler for the transition to a circular economy, using the 10R framework to explore AI applications across circularity levels. It highlights how AI can identify opportunities for material reduction, pr…

Parchami J, Darroudi A, Ali A +2

This paper proposes a hybrid framework combining Variational Mode Decomposition (VMD) with an Adaptive Learning Rate Enhanced Gated Recurrent Unit (ALRE-GRU) optimized by the Enhanced Dynamic Grasshopper Optimization Algorithm (EDGOA) for p…

(著者不明)

This study explores the application of large language models (LLMs) in finance, particularly for stock prediction and ESG evaluation. It compares human judgment with LLM-based analysis, examining the impact on investment performance.

(著者不明)

This paper proposes AI-driven methods in corporate finance to enhance governance and sustainability practices. It demonstrates how AI can automate ESG evaluation and disclosure, strengthening corporate sustainability efforts.

(著者不明)

This paper proposes a human-AI collaboration framework using LLMs for creating benchmark datasets for ESG rating agencies. It achieves efficient and high-quality dataset construction, contributing to standardization in ESG evaluation.

(著者不明)

This industrial survey examines how financial institutions use NLP to extract unstructured ESG data from financial reports, news, and filings. It provides an overview of data extraction methods, value creation potential, and practical chall…

(著者不明)

This study systematically assesses corporate ESG performance using large language models (LLMs) and examines its relationship with stock prices. It demonstrates the applicability of LLMs in ESG analysis.

(著者不明)

This paper proposes a methodology for automating the analysis of ESG disclosures in the oil and gas sector. It demonstrates scalability and adaptability for extracting insights from large-scale disclosure documents, highlighting potential f…

(著者不明)

This paper proposes a method to detect greenwashing by leveraging LLMs to analyze the gap between ESG disclosure and performance scores. Using large language models for text analysis of corporate sustainability reports, it identifies false …

Peng Gao, Kunpeng Zhang, Zongchuan Liu

This paper uses double machine learning to measure urban energy rebound effects (ERE) in China and evaluates the impact of dual-pilot policies (National Big Data Comprehensive Experimental Zones and New Energy Demonstration Cities). Results…

O. Issah, Mutala Zubeiru, Samuel Anaba

This study uses machine learning (Random Forest, XGBoost) on a global sample of 5,000 firms across 50 countries to predict ESG reporting quality. XGBoost achieves R² of 0.78 vs 0.62 for panel regression. SHAP analysis identifies firm size, …

Xiaonan Dong, sungjin son

This study examines how AI adoption affects corporate carbon neutrality performance using Resource Orchestration Theory. Analyzing panel data of Chinese A-share listed manufacturing firms (2018-2023), it finds that AI significantly enhances…

A. Sunitha, K. Srinivas, T.Radhika, B.Chandrakala Naik, P. Sandya Rani

This study examines how AI, big data analytics, and FinTech innovations affect ESG disclosure quality and corporate transparency, using survey data from 312 professionals in India, UAE, and UK, analyzed via PLS-SEM. Results show all three d…

A. Sunitha, K. Srinivas, T.Radhika, B.Chandrakala Naik, P. Sandya Rani

This paper empirically examines the combined impact of AI, big data, and fintech on ESG disclosure quality and corporate transparency using PLS-SEM on a sample of 312 professionals from India, UAE, and UK. Results show all three digital con…

Abdullah Albizri, Ahmad Jumah

This study develops a large language model (LLM) approach to measure firms' conformity with the TCFD framework from unstructured sustainability reports. Analyzing a sample of U.S.-listed firms, it finds that higher TCFD conformity is associ…