Mengze Hong
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Mengze Hong has a total of 25 co-authors. Here are the top co-authors by number of shared publications:
- Chen Jason Zhang
The Hong Kong Polytechnic University, Hong Kong, Hong Kong
3 shared publications - Kun Zhang
School of Computer Science, University of Nottingham Malaysia, Malaysia
3 shared publications - Baicheng Lin
Department of Education, Sejong University, Seoul, Republic of Korea
2 shared publications - Di Jiang
Hong Kong Polytechnic University, Hong Kong, China
2 shared publications - Hyo-Jeong So
Ewha Womans University
2 shared publications - Lingxi Jin
Ewha Womans University
2 shared publications
Latest Publications
- Open Access
- 28 May 2026
- Published by Association for Computing Machinery (ACM)
- p. 97-100
Multimodal Peer Review Simulation with Actionable To-Do Recommendations for Community-Aware Manuscript Revisions
- 28 May 2026
- Published by Association for Computing Machinery (ACM)
- p. 97-100
Open AccessMultimodal Peer Review Simulation with Actionable To-Do Recommendations for Community-Aware Manuscript Revisions
- Open Access
- 28 May 2026
- Published by Association for Computing Machinery (ACM)
- p. 132-135
CiteLLM: An Agentic Platform for Trustworthy Scientific Reference Discovery
- 28 May 2026
- Published by Association for Computing Machinery (ACM)
- p. 132-135
Open AccessCiteLLM: An Agentic Platform for Trustworthy Scientific Reference Discovery
- Open Access
- 13 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-6
How Learners Engage with an LLM-Based Pedagogical Conversational Agent During Music Form Analysis
- 13 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-6
Open AccessHow Learners Engage with an LLM-Based Pedagogical Conversational Agent During Music Form Analysis
- 21 July 2025
- Published by Springer Nature
- p. 148-155
Learning by Teaching: Enhancing Music Learning Through LLM-Based Teachable Agents
- 21 July 2025
- Published by Springer Nature
- p. 148-155
Learning by Teaching: Enhancing Music Learning Through LLM-Based Teachable Agents
- Open Access
Technical Report: A Practical Guide to Kaldi ASR Optimization
This technical report introduces innovative optimizations for Kaldi-based Automatic Speech Recognition (ASR) systems, focusing on acoustic model enhancement, hyperparameter tuning, and language model efficiency. We developed a custom Conformer block integrated with a multistream TDNN-F structure, enabling superior feature extraction and temporal modeling. Our approach includes advanced data augmentation techniques and dynamic hyperparameter optimization to boost performance and reduce overfitting. Additionally, we propose robust strategies for language model management, employing Bayesian optimization and $n$-gram pruning to ensure relevance and computational efficiency. These systematic improvements significantly elevate ASR accuracy and robustness, outperforming existing methods and offering a scalable solution for diverse speech recognition scenarios. This report underscores the importance of strategic optimizations in maintaining Kaldi's adaptability and competitiveness in rapidly evolving technological landscapes.
Open AccessTechnical Report: A Practical Guide to Kaldi ASR Optimization
This technical report introduces innovative optimizations for Kaldi-based Automatic Speech Recognition (ASR) systems, focusing on acoustic model enhancement, hyperparameter tuning, and language model efficiency. We developed a custom Conformer block integrated with a multistream TDNN-F structure, enabling superior feature extraction and temporal modeling. Our approach includes advanced data augmentation techniques and dynamic hyperparameter optimization to boost performance and reduce overfitting. Additionally, we propose robust strategies for language model management, employing Bayesian optimization and $n$-gram pruning to ensure relevance and computational efficiency. These systematic improvements significantly elevate ASR accuracy and robustness, outperforming existing methods and offering a scalable solution for diverse speech recognition scenarios. This report underscores the importance of strategic optimizations in maintaining Kaldi's adaptability and competitiveness in rapidly evolving technological landscapes.
- Open Access
- 28 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
LLM-in-the-Loop: Replicating Human Insight with LLMs for Better Machine Learning Applications
Building on the success of human-in-the-loop, where human wisdom is integrated into the development of machine learning algorithms, we take the initiative to envision an innovative and promising paradigm, LLM-in-the-loop (LLM-ITL) , which leverages the unique advantages of LLMs to replicate human involvement and offer a more flexible and cost-efficient solution to real-world challenges. Through a comprehensive review of LLM research from 2020 to 2025, we reveal that many existing LLM applications inherently align with LLM-ITL, with researchers rapidly claiming their superiority over machine learning baselines and LLM-native solutions; however, no universal definition exists, hindering its further advancement and application. In this paper, we define and categorize LLM-ITL methodologies for data, model, and task-centric applications, discuss their underlying rationale, and highlight emerging areas where LLMs can be further integrated into the loop. Furthermore, we present opportunities for developing better LLM-ITL solutions with technical advancements, such as LLM crowdsourcing and text-to-solution, establishing the proposed paradigm as a promising avenue for the future of LLM applications and machine learning research.
- 28 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
Open AccessLLM-in-the-Loop: Replicating Human Insight with LLMs for Better Machine Learning Applications
Building on the success of human-in-the-loop, where human wisdom is integrated into the development of machine learning algorithms, we take the initiative to envision an innovative and promising paradigm, LLM-in-the-loop (LLM-ITL) , which leverages the unique advantages of LLMs to replicate human involvement and offer a more flexible and cost-efficient solution to real-world challenges. Through a comprehensive review of LLM research from 2020 to 2025, we reveal that many existing LLM applications inherently align with LLM-ITL, with researchers rapidly claiming their superiority over machine learning baselines and LLM-native solutions; however, no universal definition exists, hindering its further advancement and application. In this paper, we define and categorize LLM-ITL methodologies for data, model, and task-centric applications, discuss their underlying rationale, and highlight emerging areas where LLMs can be further integrated into the loop. Furthermore, we present opportunities for developing better LLM-ITL solutions with technical advancements, such as LLM crowdsourcing and text-to-solution, establishing the proposed paradigm as a promising avenue for the future of LLM applications and machine learning research.
- Open Access
Auto-Demo Prompting: Leveraging Generated Outputs as Demonstrations for Enhanced Batch Prompting
Batch prompting is a common technique in large language models (LLMs) used to process multiple inputs simultaneously, aiming to improve computational efficiency. However, as batch sizes increase, performance degradation often occurs due to the model's difficulty in handling lengthy context inputs. Existing methods that attempt to mitigate these issues rely solely on batch data arrangement and majority voting rather than improving the design of the batch prompt itself. In this paper, we address these limitations by proposing "Auto-Demo Prompting," a novel approach that leverages the question-output pairs from earlier questions within a batch as demonstrations for subsequent answer inference. We provide a formal theoretical analysis of how Auto-Demo Prompting functions within the autoregressive generation process of LLMs, illustrating how it utilizes prior outputs to optimize the model's internal representations. Our method effectively bridges the gap between batch prompting and few-shot prompting, enhancing performance with only a slight compromise in token usage. Experimental results across five NLP tasks demonstrate its effectiveness in mitigating performance degradation and occasionally outperforming single prompts. Furthermore, it opens new avenues for applying few-shot learning techniques, such as demonstration selection, within batch prompting, making it a robust solution for real-world applications.
Open AccessAuto-Demo Prompting: Leveraging Generated Outputs as Demonstrations for Enhanced Batch Prompting
Batch prompting is a common technique in large language models (LLMs) used to process multiple inputs simultaneously, aiming to improve computational efficiency. However, as batch sizes increase, performance degradation often occurs due to the model's difficulty in handling lengthy context inputs. Existing methods that attempt to mitigate these issues rely solely on batch data arrangement and majority voting rather than improving the design of the batch prompt itself. In this paper, we address these limitations by proposing "Auto-Demo Prompting," a novel approach that leverages the question-output pairs from earlier questions within a batch as demonstrations for subsequent answer inference. We provide a formal theoretical analysis of how Auto-Demo Prompting functions within the autoregressive generation process of LLMs, illustrating how it utilizes prior outputs to optimize the model's internal representations. Our method effectively bridges the gap between batch prompting and few-shot prompting, enhancing performance with only a slight compromise in token usage. Experimental results across five NLP tasks demonstrate its effectiveness in mitigating performance degradation and occasionally outperforming single prompts. Furthermore, it opens new avenues for applying few-shot learning techniques, such as demonstration selection, within batch prompting, making it a robust solution for real-world applications.
- Open Access
- 27 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 10-16
A Comparative Study of LSTM, LightGBM, and Autoregressive Model in Narrow-Based ETF Market Prediction with Multi-Ticker Models
Narrow-based Exchange-Traded Funds (ETFs) offer significant diversification compared to stock market investments while maintaining a reasonable level of predictability through asset combinations and industry dependencies, rendering them ideal for the study of financial forecasting. In this research, a comparative analysis is presented focusing on Machine Learning (LightGBM), Deep Learning (LSTM), and Statistical (AR) methods for predicting the ETF market trends, with the autoregressive (AR) model outperforming across various evaluation metrics (RMSE, MAPE, R2). Beyond comparing different modelling techniques, the data aggregation approaches for constructing a multi-ticker model were discussed. The LSTM was greatly improved by the simple data concatenation, allowing complex training with a larger dataset, while the proposed shuffled-concatenation technique exhibits superior performance in LightGBM, surpassing all other models. These findings highlight the effectiveness of the multi-ticker model in improving prediction accuracy and emphasize the feasibility of forecasting shuffled time-series data using the tree-based model.
- 27 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 10-16
Open AccessA Comparative Study of LSTM, LightGBM, and Autoregressive Model in Narrow-Based ETF Market Prediction with Multi-Ticker Models
Narrow-based Exchange-Traded Funds (ETFs) offer significant diversification compared to stock market investments while maintaining a reasonable level of predictability through asset combinations and industry dependencies, rendering them ideal for the study of financial forecasting. In this research, a comparative analysis is presented focusing on Machine Learning (LightGBM), Deep Learning (LSTM), and Statistical (AR) methods for predicting the ETF market trends, with the autoregressive (AR) model outperforming across various evaluation metrics (RMSE, MAPE, R2). Beyond comparing different modelling techniques, the data aggregation approaches for constructing a multi-ticker model were discussed. The LSTM was greatly improved by the simple data concatenation, allowing complex training with a larger dataset, while the proposed shuffled-concatenation technique exhibits superior performance in LightGBM, surpassing all other models. These findings highlight the effectiveness of the multi-ticker model in improving prediction accuracy and emphasize the feasibility of forecasting shuffled time-series data using the tree-based model.
- Open Access
- 15 September 2023
- Published by Association for Computing Machinery (ACM)
- p. 41-47
Novel Domain-Knowledge Based Feature Selection Framework for Price Prediction: Comprehensive Modelling in Sailboat Market
- 15 September 2023
- Published by Association for Computing Machinery (ACM)
- p. 41-47
Open AccessNovel Domain-Knowledge Based Feature Selection Framework for Price Prediction: Comprehensive Modelling in Sailboat Market