GX & Decarbonization Research

Md. Abul Kalam Azad, A. Masum, M. Rahman +3

EcoStack-Pro is a federated learning framework using a stacked ensemble of LightGBM, XGBoost, and Gradient Boosting with a Fed-GenAdaptive algorithm, achieving high-precision ESG auditing across heterogeneous industrial sectors while preser…

Hye-Kyoung An, Moon-Koo Kim, Byoung-Chang Choi +1

This study classifies 5,097 ESG publications into environmental (ENV) and non-environmental (non-ENV) domains, comparing their knowledge structures using bibliometric analysis, international collaboration network analysis, and BERTopic mode…

Wenjun Wang, N. Jamil, Jun Zeng

This paper proposes a dynamic framework enhanced by text mining to analyze the impact of green finance policy on corporate sustainability. It uses literature-calibrated simulation and regularized XGBoost for validation, providing quantitati…

Aryaman Ahuja, Aryan Dutta

This paper presents an AI-enhanced digital twin approach for carbon capture and storage (CCS) under extreme climate conditions, using Bahrain as a case study. It explores how digital twin technology combined with AI can optimize CCS operati…

Yun-Kae Kiang, King H. Lam

This paper proposes ReGMS, a retrieval-grounded multi-agent architecture for climate scenario analysis to support IFRS S2/TCFD reporting. Specialized LLM agents coordinate to build and verify transition and physical-risk scenarios. Using NG…

Zahra Yaghoubi, Mahyar Mansouri, Hosein Alimardani +2

This study applies machine learning to optimize methane tri-reforming (TRM) for low-emission hydrogen production. Using 46,464 operating points, neural networks (NN), random forest, XGBoost, and others were trained, with NN achieving best p…

Chengji Liang, Yantao Li, Jianquan Guo

This paper analyzes the effect of green technology innovation on low-carbon transformation in the logistics industry using panel data and machine learning. Findings indicate that specific technologies significantly contribute to emissions r…

Mohsin Khan¹, Wendy Ashurst²

This paper proposes an AI-enhanced governance framework to improve ESG reporting integrity. It balances algorithmic detection with human judgment in a sector-specific manner, aiming to reduce greenwashing and enhance disclosure quality.

Clément Bamogo, Mustapha Oudani, Amine Belhadi +2

This paper proposes a two-stage stochastic optimization model for fertilizer planning that supports the transition to low-emission fertilizers under cap-and-trade regulation. A reinforcement Q-learning-enhanced variable neighborhood search …

Constantinos Varsos, Adamantia Stamou, George Stamoulis +1

This presentation introduces an incentive-based framework for improving energy efficiency in ICT services, from the EXIGENCE project. It explores gamified incentives for sustainable video streaming, renewable-aware federated learning, and f…

Neetal Vyas, Manmohan Vyas, Komal Singh

This paper examines how fintech startups leverage blockchain, AI, and machine learning to drive sustainable finance, including ESG analysis, green payments, and carbon credits. They enhance transparency, reduce greenwashing, and fill gaps l…

Ameni Boumaiza

This paper proposes an AI-driven demand-side management approach for sustainable power systems in Qatar, integrating reliability and carbon pricing to achieve decarbonization and grid stability.

Dauda Bola Abdulsalam, Duminda Kuruppuarachchi, Helen Roberts

This study applies NLP to ESG/sustainability reports of S&P 500 firms (2018-2023) to measure TNFD framework adherence. Findings show that TNFD alignment, clear nature-related risk/opportunity identification, and comprehensive reporting are …

Ngone Mirimi, H. Manuere

AI-enabled climate decision-support systems are de facto governance infrastructures with political and accountability consequences, yet they are treated as neutral technical tools. The paper reframes transparency as an enforceable instituti…

Dr. Thanakit Ouanhlee

This study examines the relationship between AI integration in compensation systems and human capital accounting disclosure (HCAD) quality among manufacturing organizations in Thailand. It finds that while AI integration is moderate, disclo…

Adrian Lim, Putri Rahayu

This review synthesizes AI applications across financial time-series forecasting, portfolio construction, and firm-level sustainability analysis, with emphasis on AI-driven ESG rating prediction and disclosure signal extraction. It argues f…

Christie Azour

This paper demonstrates AI's effectiveness in ESG investing. AI processed ESG documents 1,200 times faster than humans, reduced rating divergence by 34%, detected greenwashing with 83% accuracy, identified climate risks 8-14 months earlier,…

Hasan Talaş, E. Gök, Özen Akçakanat +4

This study uses tree-based machine learning on 428 Turkish non-financial firms to show that ESG and governance signals provide statistically significant additional information for predicting ROE beyond traditional financial ratios. Random F…

Cristina Caro-González, Daniel Jato-Espino, Yudith Cardinale

This systematic review of 64 studies on integrating ESG factors into financial risk prediction finds that traditional econometrics still dominate (48%), but ML (39%), NLP (8%) are growing. ML models, especially ensemble methods and neural n…

Chen Xu, Batkhuyag Ganbaatar

This study examines the impact of intelligent manufacturing on ESG performance for Chinese A-share listed companies (2015-2023) using two-way fixed-effects panel regression, PSM-DID, and System GMM. It finds that intelligent manufacturing s…