Haohui Lu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Haohui Lu has a total of 43 co-authors. Here are the top co-authors by number of shared publications:
- Shahadat Uddin
School of Electrical and Computer Engineering, The University of Sydney, Darlington, Australia
14 shared publications - Usman Naseem
Macquarie University, Australia
3 shared publications - Arif Khan
The University of Sydney
2 shared publications - Fangyu Zhou
School of Electrical and Computer Engineering, The University of Sydney, Sydney, NSW, Australia
2 shared publications - Farshid Hajati
University of New England
2 shared publications - Junbin Gao
University of Sydney
2 shared publications
Latest Publications
- Open Access
- 31 March 2026
- Published by Springer Nature in Machine Learning
- Vol. 115 (4) , 79
Bias Mitigation in Large Language Models for Tabular Data Classification
Large Language Models (LLMs) perform well in tabular prediction tasks with limited data, using their ability to understand instructions and learn from examples. However, their reliance on training data can perpetuate social biases, leading to unfair outcomes and disproportionately impacting underprivileged groups. Addressing these biases is critical as LLMs see wider adoption in tabular data tasks. Traditional bias mitigation strategies in machine learning, such as balancing datasets or applying fairness constraints, are less effective with LLMs. Our research explores whether bias in LLMs for tabular data classification can be mitigated. Through extensive experiments, we found that using LLMs in a zero-shot setting introduces bias, and in-context learning slightly reduces these disparities. Meanwhile, fine-tuning and retrieval augmented generation show limited effectiveness in bias mitigation. We introduced three instruction-based prompting strategies to enhance fairness: Fair Prompting , Generalised Prompting , and Descriptive Prompting . The results show that combining descriptive prompting with in-context learning, particularly the Equal Samples Across Demographics approach, consistently narrowed fairness gaps across demographic subgroups, and yielded accuracy gains ranging from 3.27% to 15.05% across multiple datasets, underscoring its potential as a promising strategy in the ongoing effort to mitigate bias in LLMs.
- 31 March 2026
- Published by Springer Nature in Machine Learning
- Vol. 115 (4) , 79
Open AccessBias Mitigation in Large Language Models for Tabular Data Classification
Large Language Models (LLMs) perform well in tabular prediction tasks with limited data, using their ability to understand instructions and learn from examples. However, their reliance on training data can perpetuate social biases, leading to unfair outcomes and disproportionately impacting underprivileged groups. Addressing these biases is critical as LLMs see wider adoption in tabular data tasks. Traditional bias mitigation strategies in machine learning, such as balancing datasets or applying fairness constraints, are less effective with LLMs. Our research explores whether bias in LLMs for tabular data classification can be mitigated. Through extensive experiments, we found that using LLMs in a zero-shot setting introduces bias, and in-context learning slightly reduces these disparities. Meanwhile, fine-tuning and retrieval augmented generation show limited effectiveness in bias mitigation. We introduced three instruction-based prompting strategies to enhance fairness: Fair Prompting , Generalised Prompting , and Descriptive Prompting . The results show that combining descriptive prompting with in-context learning, particularly the Equal Samples Across Demographics approach, consistently narrowed fairness gaps across demographic subgroups, and yielded accuracy gains ranging from 3.27% to 15.05% across multiple datasets, underscoring its potential as a promising strategy in the ongoing effort to mitigate bias in LLMs.
- 1 December 2025
- Published by Springer Nature in AI and Ethics
- Vol. 6 (1) , 25
Fairness-preserving framework for machine learning: data bias quantification, model evaluation, and robustness across multiple datasets
- 1 December 2025
- Published by Springer Nature in AI and Ethics
- Vol. 6 (1) , 25
Fairness-preserving framework for machine learning: data bias quantification, model evaluation, and robustness across multiple datasets
- 27 November 2025
- Published by Springer Nature
- p. 335-346
An Explainable Graph Learning Framework for Severe Maternal Morbidity Prediction
- 27 November 2025
- Published by Springer Nature
- p. 335-346
An Explainable Graph Learning Framework for Severe Maternal Morbidity Prediction
- 25 November 2025
- Published by Springer Nature
- p. 163-176
Assessing Algorithmic Fairness in Socioeconomic Predictions Using Australian Census Data
- 25 November 2025
- Published by Springer Nature
- p. 163-176
Assessing Algorithmic Fairness in Socioeconomic Predictions Using Australian Census Data
- Open Access
- 16 July 2025
- Published by Elsevier in Artificial Intelligence in Medicine
- Vol. 168, 103216
Toward fair medical advice: Addressing and mitigating bias in large language model-based healthcare applications
- 16 July 2025
- Published by Elsevier in Artificial Intelligence in Medicine
- Vol. 168, 103216
Open AccessToward fair medical advice: Addressing and mitigating bias in large language model-based healthcare applications
- Open Access
- 8 May 2025
- Published by Association for Computing Machinery (ACM)
- p. 2071-2077
KG-UQ: Knowledge Graph-Based Uncertainty Quantification for Long Text in Large Language Models
- 8 May 2025
- Published by Association for Computing Machinery (ACM)
- p. 2071-2077
Open AccessKG-UQ: Knowledge Graph-Based Uncertainty Quantification for Long Text in Large Language Models
- Open Access
- 18 April 2025
- Published by Springer Nature in Research Square
Fairness-Preserving Implementation of Machine Learning Models
Fairness in machine learning systems is essential for building trustworthy, ethical, and socially responsible AI, particularly in high-stakes domains such as healthcare and human services. This study proposes a comprehensive fairness-preserving framework integrating data bias quantification with model-level fairness evaluation and eliminating its violation. The framework uses Earth Mover’s Distance to quantify the distributional discrepancy between subgroups and the overall population, providing a statistical foundation for identifying group-level data bias. We assess fairness across five widely accepted definitions (i.e., demographic parity, equalised odds, equal opportunity, false positive rate parity, and predictive parity), each derived from the confusion matrix outcomes of ML models. The framework is empirically validated using a real-world health dataset and five commonly used supervised learning algorithms: Logistic Regression, Decision Tree, Random Forest, Support Vector Machine, and K-Nearest Neighbours. Results show that fairness-preserving adjustments, mainly through targeted data modification, significantly reduce fairness violations with minimal impact on overall model performance. By combining data-level bias quantification with robust statistical validation, this work offers a practical and interpretable approach to implementing fairness in ML systems. The framework lays a foundation for future extensions incorporating intersectional fairness, multi-class classification, and dynamic data environments. It contributes toward the development of AI systems that are not only accurate but also equitable and accountable.
- 18 April 2025
- Published by Springer Nature in Research Square
Open AccessFairness-Preserving Implementation of Machine Learning Models
Fairness in machine learning systems is essential for building trustworthy, ethical, and socially responsible AI, particularly in high-stakes domains such as healthcare and human services. This study proposes a comprehensive fairness-preserving framework integrating data bias quantification with model-level fairness evaluation and eliminating its violation. The framework uses Earth Mover’s Distance to quantify the distributional discrepancy between subgroups and the overall population, providing a statistical foundation for identifying group-level data bias. We assess fairness across five widely accepted definitions (i.e., demographic parity, equalised odds, equal opportunity, false positive rate parity, and predictive parity), each derived from the confusion matrix outcomes of ML models. The framework is empirically validated using a real-world health dataset and five commonly used supervised learning algorithms: Logistic Regression, Decision Tree, Random Forest, Support Vector Machine, and K-Nearest Neighbours. Results show that fairness-preserving adjustments, mainly through targeted data modification, significantly reduce fairness violations with minimal impact on overall model performance. By combining data-level bias quantification with robust statistical validation, this work offers a practical and interpretable approach to implementing fairness in ML systems. The framework lays a foundation for future extensions incorporating intersectional fairness, multi-class classification, and dynamic data environments. It contributes toward the development of AI systems that are not only accurate but also equitable and accountable.
- Open Access
- 5 December 2024
- Published by Springer Nature in Journal of Computational Social Science
- Vol. 8 (1) , 1-21
Identifying the factors influencing the development of bilateral investment treaties with health safeguards: a Machine Learning-based link prediction approach
A network analysis approach, complemented by machine learning (ML) techniques, is applied to analyse the factors influencing Bilateral Investment Treaties (BITs) at the country level. Using the Electronic Database of Investment Treaties, BITs with health safeguards from 167 countries were charted, resulting in 534 connections with countries as nodes and their BITs as edges. Network analysis found that, on average, a country established BITs with six other nations. Additionally, we used node embedding techniques to generate features from the network, such as the Jaccard coefficient, resource allocation, and Adamic Adar for downstream link prediction. This study employed five tree-based ML models to predict future BIT formations with health inclusion. The eXtreme Gradient Boosting model proved to be superior, achieving a 64.02% accuracy rate. Notably, the Common Neighbor centrality feature and the Capital Account Balance Ratio emerged as influential factors in creating new BITs with health inclusions. Beyond economic considerations, our study highlighted a vital intersection: the nexus between BITs, economic growth, and public health policies. In essence, this research underscores the importance of safeguarding public health in BITs and showcases the potential of ML in understanding the intricacies of international treaties.
- 5 December 2024
- Published by Springer Nature in Journal of Computational Social Science
- Vol. 8 (1) , 1-21
Open AccessIdentifying the factors influencing the development of bilateral investment treaties with health safeguards: a Machine Learning-based link prediction approach
A network analysis approach, complemented by machine learning (ML) techniques, is applied to analyse the factors influencing Bilateral Investment Treaties (BITs) at the country level. Using the Electronic Database of Investment Treaties, BITs with health safeguards from 167 countries were charted, resulting in 534 connections with countries as nodes and their BITs as edges. Network analysis found that, on average, a country established BITs with six other nations. Additionally, we used node embedding techniques to generate features from the network, such as the Jaccard coefficient, resource allocation, and Adamic Adar for downstream link prediction. This study employed five tree-based ML models to predict future BIT formations with health inclusion. The eXtreme Gradient Boosting model proved to be superior, achieving a 64.02% accuracy rate. Notably, the Common Neighbor centrality feature and the Capital Account Balance Ratio emerged as influential factors in creating new BITs with health inclusions. Beyond economic considerations, our study highlighted a vital intersection: the nexus between BITs, economic growth, and public health policies. In essence, this research underscores the importance of safeguarding public health in BITs and showcases the potential of ML in understanding the intricacies of international treaties.
- Open Access
STLLM-DF: A Spatial-Temporal Large Language Model with Diffusion for Enhanced Multi-Mode Traffic System Forecasting
The rapid advancement of Intelligent Transportation Systems (ITS) presents challenges, particularly with missing data in multi-modal transportation and the complexity of handling diverse sequential tasks within a centralized framework. To address these issues, we propose the Spatial-Temporal Large Language Model Diffusion (STLLM-DF), an innovative model that leverages Denoising Diffusion Probabilistic Models (DDPMs) and Large Language Models (LLMs) to improve multi-task transportation prediction. The DDPM's robust denoising capabilities enable it to recover underlying data patterns from noisy inputs, making it particularly effective in complex transportation systems. Meanwhile, the non-pretrained LLM dynamically adapts to spatial-temporal relationships within multi-modal networks, allowing the system to efficiently manage diverse transportation tasks in both long-term and short-term predictions. Extensive experiments demonstrate that STLLM-DF consistently outperforms existing models, achieving an average reduction of 2.40\% in MAE, 4.50\% in RMSE, and 1.51\% in MAPE. This model significantly advances centralized ITS by enhancing predictive accuracy, robustness, and overall system performance across multiple tasks, thus paving the way for more effective spatio-temporal traffic forecasting through the integration of frozen transformer language models and diffusion techniques.
Open AccessSTLLM-DF: A Spatial-Temporal Large Language Model with Diffusion for Enhanced Multi-Mode Traffic System Forecasting
The rapid advancement of Intelligent Transportation Systems (ITS) presents challenges, particularly with missing data in multi-modal transportation and the complexity of handling diverse sequential tasks within a centralized framework. To address these issues, we propose the Spatial-Temporal Large Language Model Diffusion (STLLM-DF), an innovative model that leverages Denoising Diffusion Probabilistic Models (DDPMs) and Large Language Models (LLMs) to improve multi-task transportation prediction. The DDPM's robust denoising capabilities enable it to recover underlying data patterns from noisy inputs, making it particularly effective in complex transportation systems. Meanwhile, the non-pretrained LLM dynamically adapts to spatial-temporal relationships within multi-modal networks, allowing the system to efficiently manage diverse transportation tasks in both long-term and short-term predictions. Extensive experiments demonstrate that STLLM-DF consistently outperforms existing models, achieving an average reduction of 2.40\% in MAE, 4.50\% in RMSE, and 1.51\% in MAPE. This model significantly advances centralized ITS by enhancing predictive accuracy, robustness, and overall system performance across multiple tasks, thus paving the way for more effective spatio-temporal traffic forecasting through the integration of frozen transformer language models and diffusion techniques.