Haopeng Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Haopeng Zhang has a total of 12 co-authors. Here are the top co-authors by number of shared publications:
- Abdur Rasool
Department of Information and Computer Sciences, University of Hawaii at Manoa, Honolulu, HI 96822, USA; Department of Statistics and Data Science, Southern University of Science and Technology , Shenzhen, 518055, China, USA
2 shared publications - Dennis Paul Wall
Department of Pediatrics, Division of Clinical Informatics, Stanford University School of Medicine, Palo Alto, CA 94305, USA
2 shared publications - Haohan Yuan
University of Hawaii at Manoa, Honolulu, USA
2 shared publications - Jingzhe Zhang
University of South Florida, Tampa, USA
2 shared publications - Kaiying Lin
University of Hawai’i at Manoa
2 shared publications - Onur Cezmi Mutlu
Department of Biomedical Data Science and Department of Pediatrics, Stanford University, Stanford, CA 94305, USA
2 shared publications
Latest Publications
- Open Access
- 21 October 2025
- Published by JMIR Publications Inc. in JMIR AI
- Vol. 4, e75030
Aiding Large Language Models Using Clinical Scoresheets for Neurobehavioral Diagnostic Classification From Text: Algorithm Development and Validation
Background: Large language models (LLMs) have demonstrated the ability to perform complex tasks traditionally requiring human intelligence. However, their use in automated diagnostics for psychiatry and behavioral sciences remains under-studied. Objective: This study aimed to evaluate whether incorporating structured clinical assessment scales improved the diagnostic performance of LLM-based chatbots for neuropsychiatric conditions (we evaluated autism spectrum disorder, aphasia, and depression datasets) across two prompting strategies: (1) direct diagnosis and (2) code generation. We aimed to contextualize LLM-based diagnostic performance by benchmarking it against prior work that applied traditional machine learning classifiers to the same datasets, allowing us to assess whether LLMs offer competitive or complementary capabilities in clinical classification tasks. Methods: We tested two approaches using ChatGPT, Gemini, and Claude models: (1) direct diagnostic querying and (2) execution of chatbot-generated code for classification. Three diagnostic datasets were used: ASDBank (autism spectrum disorder), AphasiaBank (aphasia), and Distress Analysis Interview Corpus-Wizard-of-Oz interviews (depression and related conditions). Each approach was evaluated with and without the aid of clinical assessment scales. Performance was compared to existing machine learning benchmarks on these datasets. Results: Across all 3 datasets, incorporating clinical assessment scales led to little improvement in performance, and results remained inconsistent and generally below those reported in previous studies. On the AphasiaBank dataset, the direct diagnosis approach using ChatGPT with GPT-4 produced a low F1-score of 65.6% and specificity of 33%. The code generation method improved results, with ChatGPT with GPT-4o reaching an F1-score of 81.4%, specificity of 78.6%, and sensitivity of 84.3%. ChatGPT with GPT-o3 and Gemini 2.5 Pro performed even better, with F1-scores of 86.5% and 84.3%, respectively. For the ASDBank dataset, direct diagnosis results were lower, with F1-scores of 56% for ChatGPT with GPT-4 and 54% for ChatGPT with GPT-4o. Under code generation, ChatGPT with GPT-o3 reached 67.9%, and Claude 3.5 performed reasonably well with 60%. Gemini 2.5 Pro failed to respond under this assessment condition. In the Distress Analysis Interview Corpus-Wizard-of-Oz dataset, direct diagnosis yielded high accuracy (70.9%) but poor F1-scores of 8% using ChatGPT with GPT-4o. Code generation improved specificity—88.6% with ChatGPT with GPT-4o—but F1-scores remained low overall. These findings suggest that, while clinical scales may help structure outputs, prompting alone remains insufficient for consistent diagnostic accuracy. Conclusions: Current LLM-based chatbots, when prompted naively, underperform on psychiatric and behavioral diagnostic tasks compared to specialized machine learning models. Clinical assessment scales might modestly aid chatbot performance, but more sophisticated prompt engineering and domain integration are likely required to reach clinically actionable standards.
- 21 October 2025
- Published by JMIR Publications Inc. in JMIR AI
- Vol. 4, e75030
Open AccessAiding Large Language Models Using Clinical Scoresheets for Neurobehavioral Diagnostic Classification From Text: Algorithm Development and Validation
Background: Large language models (LLMs) have demonstrated the ability to perform complex tasks traditionally requiring human intelligence. However, their use in automated diagnostics for psychiatry and behavioral sciences remains under-studied. Objective: This study aimed to evaluate whether incorporating structured clinical assessment scales improved the diagnostic performance of LLM-based chatbots for neuropsychiatric conditions (we evaluated autism spectrum disorder, aphasia, and depression datasets) across two prompting strategies: (1) direct diagnosis and (2) code generation. We aimed to contextualize LLM-based diagnostic performance by benchmarking it against prior work that applied traditional machine learning classifiers to the same datasets, allowing us to assess whether LLMs offer competitive or complementary capabilities in clinical classification tasks. Methods: We tested two approaches using ChatGPT, Gemini, and Claude models: (1) direct diagnostic querying and (2) execution of chatbot-generated code for classification. Three diagnostic datasets were used: ASDBank (autism spectrum disorder), AphasiaBank (aphasia), and Distress Analysis Interview Corpus-Wizard-of-Oz interviews (depression and related conditions). Each approach was evaluated with and without the aid of clinical assessment scales. Performance was compared to existing machine learning benchmarks on these datasets. Results: Across all 3 datasets, incorporating clinical assessment scales led to little improvement in performance, and results remained inconsistent and generally below those reported in previous studies. On the AphasiaBank dataset, the direct diagnosis approach using ChatGPT with GPT-4 produced a low F1-score of 65.6% and specificity of 33%. The code generation method improved results, with ChatGPT with GPT-4o reaching an F1-score of 81.4%, specificity of 78.6%, and sensitivity of 84.3%. ChatGPT with GPT-o3 and Gemini 2.5 Pro performed even better, with F1-scores of 86.5% and 84.3%, respectively. For the ASDBank dataset, direct diagnosis results were lower, with F1-scores of 56% for ChatGPT with GPT-4 and 54% for ChatGPT with GPT-4o. Under code generation, ChatGPT with GPT-o3 reached 67.9%, and Claude 3.5 performed reasonably well with 60%. Gemini 2.5 Pro failed to respond under this assessment condition. In the Distress Analysis Interview Corpus-Wizard-of-Oz dataset, direct diagnosis yielded high accuracy (70.9%) but poor F1-scores of 8% using ChatGPT with GPT-4o. Code generation improved specificity—88.6% with ChatGPT with GPT-4o—but F1-scores remained low overall. These findings suggest that, while clinical scales may help structure outputs, prompting alone remains insufficient for consistent diagnostic accuracy. Conclusions: Current LLM-based chatbots, when prompted naively, underperform on psychiatric and behavioral diagnostic tasks compared to specialized machine learning models. Clinical assessment scales might modestly aid chatbot performance, but more sophisticated prompt engineering and domain integration are likely required to reach clinically actionable standards.
- Open Access
- 4 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
Bridging Multimodal and Video Summarization: A Unified Survey
Multimodal summarization (MMS) and video summarization (VS) have traditionally evolved in separate communities-natural language processing (NLP) and computer vision (CV), respectively. MMS focuses on generating textual summaries from inputs such as text, images, or audio, while VS emphasizes selecting key visual content. With the recent rise of vision-language models (VLMs), these oncedisparate tasks are converging under a unified framework that integrates visual and linguistic understanding. In this survey, we provide a unified perspective that bridges MMS and VS. We formalize the task landscape, review key datasets and evaluation metrics, and categorize major modeling approaches into new taxonomy. In addition, we highlight core challenges and outline future directions toward building general-purpose multimodal summarization systems. By synthesizing insights from both NLP and CV communities, this survey aims to establish a coherent foundation for advancing this rapidly evolving field.
- 4 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
Open AccessBridging Multimodal and Video Summarization: A Unified Survey
Multimodal summarization (MMS) and video summarization (VS) have traditionally evolved in separate communities-natural language processing (NLP) and computer vision (CV), respectively. MMS focuses on generating textual summaries from inputs such as text, images, or audio, while VS emphasizes selecting key visual content. With the recent rise of vision-language models (VLMs), these oncedisparate tasks are converging under a unified framework that integrates visual and linguistic understanding. In this survey, we provide a unified perspective that bridges MMS and VS. We formalize the task landscape, review key datasets and evaluation metrics, and categorize major modeling approaches into new taxonomy. In addition, we highlight core challenges and outline future directions toward building general-purpose multimodal summarization systems. By synthesizing insights from both NLP and CV communities, this survey aims to establish a coherent foundation for advancing this rapidly evolving field.
- Open Access
- 23 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 583-584
Poster: Large Language Model-powered Wi-Fi-based Human Activity Recognition
- 23 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 583-584
Open AccessPoster: Large Language Model-powered Wi-Fi-based Human Activity Recognition
- Open Access
- 23 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 8-13
Wi-Chat: Large Language Model-powered Wi-Fi-based Human Activity Recognition
- 23 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 8-13
Open AccessWi-Chat: Large Language Model-powered Wi-Fi-based Human Activity Recognition
- Top Cited PapersOpen Access
- 19 April 2025
- Published by Association for Computing Machinery (ACM) in ACM Computing Surveys
A Systematic Survey of Text Summarization: From Statistical Methods to Large Language Models
Text summarization research has undergone several significant transformations with the advent of deep neural networks, pre-trained language models (PLMs), and recent large language models (LLMs). This survey thus provides a comprehensive review of the research progress and evolution in text summarization through the lens of these paradigm shifts. It is organized into two main parts: (1) a detailed overview of datasets, evaluation metrics, and summarization methods before the LLM era, encompassing traditional statistical methods, deep learning approaches, and PLM fine-tuning techniques, and (2) the first detailed examination of recent advancements in benchmarking, modeling, and evaluating summarization in the LLM era. By synthesizing existing literature and presenting a cohesive overview, this survey also discusses research trends, open challenges, and proposes promising research directions in summarization, aiming to guide researchers through the evolving landscape of summarization research.
- 19 April 2025
- Published by Association for Computing Machinery (ACM) in ACM Computing Surveys
Top Cited PapersOpen AccessA Systematic Survey of Text Summarization: From Statistical Methods to Large Language Models
Text summarization research has undergone several significant transformations with the advent of deep neural networks, pre-trained language models (PLMs), and recent large language models (LLMs). This survey thus provides a comprehensive review of the research progress and evolution in text summarization through the lens of these paradigm shifts. It is organized into two main parts: (1) a detailed overview of datasets, evaluation metrics, and summarization methods before the LLM era, encompassing traditional statistical methods, deep learning approaches, and PLM fine-tuning techniques, and (2) the first detailed examination of recent advancements in benchmarking, modeling, and evaluating summarization in the LLM era. By synthesizing existing literature and presenting a cohesive overview, this survey also discusses research trends, open challenges, and proposes promising research directions in summarization, aiming to guide researchers through the evolving landscape of summarization research.
- Open Access
- 28 March 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Aiding Large Language Models Using Clinical Scoresheets for Neurobehavioral Diagnostic Classification From Text: Algorithm Development and Validation (Preprint)
BACKGROUND Large language models (LLMs) have demonstrated the ability to perform complex tasks traditionally requiring human intelligence. However, their use in automated diagnostics for psychiatry and behavioral sciences remains under-studied. OBJECTIVE This study aimed to evaluate whether incorporating structured clinical assessment scales improved the diagnostic performance of LLM-based chatbots for neuropsychiatric conditions (we evaluated autism spectrum disorder, aphasia, and depression datasets) across two prompting strategies: (1) direct diagnosis and (2) code generation. We aimed to contextualize LLM-based diagnostic performance by benchmarking it against prior work that applied traditional machine learning classifiers to the same datasets, allowing us to assess whether LLMs offer competitive or complementary capabilities in clinical classification tasks. METHODS We tested two approaches using ChatGPT, Gemini, and Claude models: (1) direct diagnostic querying and (2) execution of chatbot-generated code for classification. Three diagnostic datasets were used: ASDBank (autism spectrum disorder), AphasiaBank (aphasia), and Distress Analysis Interview Corpus-Wizard-of-Oz interviews (depression and related conditions). Each approach was evaluated with and without the aid of clinical assessment scales. Performance was compared to existing machine learning benchmarks on these datasets. RESULTS Across all 3 datasets, incorporating clinical assessment scales led to little improvement in performance, and results remained inconsistent and generally below those reported in previous studies. On the AphasiaBank dataset, the direct diagnosis approach using ChatGPT with GPT-4 produced a low F1-score of 65.6% and specificity of 33%. The code generation method improved results, with ChatGPT with GPT-4o reaching an F1-score of 81.4%, specificity of 78.6%, and sensitivity of 84.3%. ChatGPT with GPT-o3 and Gemini 2.5 Pro performed even better, with F1-scores of 86.5% and 84.3%, respectively. For the ASDBank dataset, direct diagnosis results were lower, with F1-scores of 56% for ChatGPT with GPT-4 and 54% for ChatGPT with GPT-4o. Under code generation, ChatGPT with GPT-o3 reached 67.9%, and Claude 3.5 performed reasonably well with 60%. Gemini 2.5 Pro failed to respond under this assessment condition. In the Distress Analysis Interview Corpus-Wizard-of-Oz dataset, direct diagnosis yielded high accuracy (70.9%) but poor F1-scores of 8% using ChatGPT with GPT-4o. Code generation improved specificity—88.6% with ChatGPT with GPT-4o—but F1-scores remained low overall. These findings suggest that, while clinical scales may help structure outputs, prompting alone remains insufficient for consistent diagnostic accuracy. CONCLUSIONS Current LLM-based chatbots, when prompted naively, underperform on psychiatric and behavioral diagnostic tasks compared to specialized machine learning models. Clinical assessment scales might modestly aid chatbot performance, but more sophisticated prompt engineering and domain integration are likely required to reach clinically actionable standards.
- 28 March 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessAiding Large Language Models Using Clinical Scoresheets for Neurobehavioral Diagnostic Classification From Text: Algorithm Development and Validation (Preprint)
BACKGROUND Large language models (LLMs) have demonstrated the ability to perform complex tasks traditionally requiring human intelligence. However, their use in automated diagnostics for psychiatry and behavioral sciences remains under-studied. OBJECTIVE This study aimed to evaluate whether incorporating structured clinical assessment scales improved the diagnostic performance of LLM-based chatbots for neuropsychiatric conditions (we evaluated autism spectrum disorder, aphasia, and depression datasets) across two prompting strategies: (1) direct diagnosis and (2) code generation. We aimed to contextualize LLM-based diagnostic performance by benchmarking it against prior work that applied traditional machine learning classifiers to the same datasets, allowing us to assess whether LLMs offer competitive or complementary capabilities in clinical classification tasks. METHODS We tested two approaches using ChatGPT, Gemini, and Claude models: (1) direct diagnostic querying and (2) execution of chatbot-generated code for classification. Three diagnostic datasets were used: ASDBank (autism spectrum disorder), AphasiaBank (aphasia), and Distress Analysis Interview Corpus-Wizard-of-Oz interviews (depression and related conditions). Each approach was evaluated with and without the aid of clinical assessment scales. Performance was compared to existing machine learning benchmarks on these datasets. RESULTS Across all 3 datasets, incorporating clinical assessment scales led to little improvement in performance, and results remained inconsistent and generally below those reported in previous studies. On the AphasiaBank dataset, the direct diagnosis approach using ChatGPT with GPT-4 produced a low F1-score of 65.6% and specificity of 33%. The code generation method improved results, with ChatGPT with GPT-4o reaching an F1-score of 81.4%, specificity of 78.6%, and sensitivity of 84.3%. ChatGPT with GPT-o3 and Gemini 2.5 Pro performed even better, with F1-scores of 86.5% and 84.3%, respectively. For the ASDBank dataset, direct diagnosis results were lower, with F1-scores of 56% for ChatGPT with GPT-4 and 54% for ChatGPT with GPT-4o. Under code generation, ChatGPT with GPT-o3 reached 67.9%, and Claude 3.5 performed reasonably well with 60%. Gemini 2.5 Pro failed to respond under this assessment condition. In the Distress Analysis Interview Corpus-Wizard-of-Oz dataset, direct diagnosis yielded high accuracy (70.9%) but poor F1-scores of 8% using ChatGPT with GPT-4o. Code generation improved specificity—88.6% with ChatGPT with GPT-4o—but F1-scores remained low overall. These findings suggest that, while clinical scales may help structure outputs, prompting alone remains insufficient for consistent diagnostic accuracy. CONCLUSIONS Current LLM-based chatbots, when prompted naively, underperform on psychiatric and behavioral diagnostic tasks compared to specialized machine learning models. Clinical assessment scales might modestly aid chatbot performance, but more sophisticated prompt engineering and domain integration are likely required to reach clinically actionable standards.