Hen-Hsen Huang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Hen-Hsen Huang has a total of 86 co-authors. Here are the top co-authors by number of shared publications:
- Hsin-Hsi Chen
Department of Computer Science and Information Engineering National Taiwan University Taiwan
48 shared publications - Chung-Chi Chen
Human-Agent Ally Lab, National Institute of Informatics, Tokyo, Japan
26 shared publications - An-Zi Yen
Department of Computer Science National Yang Ming Chiao Tung University Taiwan
8 shared publications - Hiroya Takamura
National Institute of Advanced Industrial Science and Technology
5 shared publications Sheng-Lun Wei
Department of Computer Science and Information Engineering National Taiwan University Taiwan
5 shared publicationsTsung-Hsuan Pan
Department of Computer Science and Information Engineering Artificial Intelligence Research Center Institute of Information Science Department of Computer Science and Information Engineering National Taiwan University Taiwan AIST Japan Academia Sinica Taiwan National Taiwan University Taiwan
3 shared publications
Latest Publications
- Open Access
A Universal Action Space for General Behavior Analysis
Analyzing animal and human behavior has long been a challenging task in computer vision. Early approaches from the 1970s to the 1990s relied on hand-crafted edge detection, segmentation, and low-level features such as color, shape, and texture to locate objects and infer their identities-an inherently ill-posed problem. Behavior analysis in this era typically proceeded by tracking identified objects over time and modeling their trajectories using sparse feature points, which further limited robustness and generalization. A major shift occurred with the introduction of ImageNet by Deng and Li in 2010, which enabled large-scale visual recognition through deep neural networks and effectively served as a comprehensive visual dictionary. This development allowed object recognition to move beyond complex low-level processing toward learned high-level representations. In this work, we follow this paradigm to build a large-scale Universal Action Space (UAS) using existing labeled human-action datasets. We then use this UAS as the foundation for analyzing and categorizing mammalian and chimpanzee behavior datasets. The source code is released on GitHub at https://github.com/franktpmvu/Universal-Action-Space.
Open AccessA Universal Action Space for General Behavior Analysis
Analyzing animal and human behavior has long been a challenging task in computer vision. Early approaches from the 1970s to the 1990s relied on hand-crafted edge detection, segmentation, and low-level features such as color, shape, and texture to locate objects and infer their identities-an inherently ill-posed problem. Behavior analysis in this era typically proceeded by tracking identified objects over time and modeling their trajectories using sparse feature points, which further limited robustness and generalization. A major shift occurred with the introduction of ImageNet by Deng and Li in 2010, which enabled large-scale visual recognition through deep neural networks and effectively served as a comprehensive visual dictionary. This development allowed object recognition to move beyond complex low-level processing toward learned high-level representations. In this work, we follow this paradigm to build a large-scale Universal Action Space (UAS) using existing labeled human-action datasets. We then use this UAS as the foundation for analyzing and categorizing mammalian and chimpanzee behavior datasets. The source code is released on GitHub at https://github.com/franktpmvu/Universal-Action-Space.
- Open Access
Diverge to Induce Prompting: Multi-Rationale Induction for Zero-Shot Reasoning
To address the instability of unguided reasoning paths in standard Chain-of-Thought prompting, recent methods guide large language models (LLMs) by first eliciting a single reasoning strategy. However, relying on just one strategy for each question can still limit performance across diverse tasks. We propose Diverge-to-Induce Prompting (DIP), a framework that first prompts an LLM to generate multiple diverse high-level rationales for each question. Each rationale is then elaborated into a detailed, step-by-step draft plan. Finally, these draft plans are induced into a final plan. DIP enhances zero-shot reasoning accuracy without reliance on resource-intensive sampling. Experiments show that DIP outperforms single-strategy prompting, demonstrating the effectiveness of multi-plan induction for prompt-based reasoning.
Open AccessDiverge to Induce Prompting: Multi-Rationale Induction for Zero-Shot Reasoning
To address the instability of unguided reasoning paths in standard Chain-of-Thought prompting, recent methods guide large language models (LLMs) by first eliciting a single reasoning strategy. However, relying on just one strategy for each question can still limit performance across diverse tasks. We propose Diverge-to-Induce Prompting (DIP), a framework that first prompts an LLM to generate multiple diverse high-level rationales for each question. Each rationale is then elaborated into a detailed, step-by-step draft plan. Finally, these draft plans are induced into a final plan. DIP enhances zero-shot reasoning accuracy without reliance on resource-intensive sampling. Experiments show that DIP outperforms single-strategy prompting, demonstrating the effectiveness of multi-plan induction for prompt-based reasoning.
- Open Access
Bias in the Ear of the Listener: Assessing Sensitivity in Audio Language Models Across Linguistic, Demographic, and Positional Variations
This work presents the first systematic investigation of speech bias in multilingual MLLMs. We construct and release the BiasInEar dataset, a speech-augmented benchmark based on Global MMLU Lite, spanning English, Chinese, and Korean, balanced by gender and accent, and totaling 70.8 hours ($\approx$4,249 minutes) of speech with 11,200 questions. Using four complementary metrics (accuracy, entropy, APES, and Fleiss' $κ$), we evaluate nine representative models under linguistic (language and accent), demographic (gender), and structural (option order) perturbations. Our findings reveal that MLLMs are relatively robust to demographic factors but highly sensitive to language and option order, suggesting that speech can amplify existing structural biases. Moreover, architectural design and reasoning strategy substantially affect robustness across languages. Overall, this study establishes a unified framework for assessing fairness and robustness in speech-integrated LLMs, bridging the gap between text- and speech-based evaluation. The resources can be found at https://github.com/ntunlplab/BiasInEar.
Open AccessBias in the Ear of the Listener: Assessing Sensitivity in Audio Language Models Across Linguistic, Demographic, and Positional Variations
This work presents the first systematic investigation of speech bias in multilingual MLLMs. We construct and release the BiasInEar dataset, a speech-augmented benchmark based on Global MMLU Lite, spanning English, Chinese, and Korean, balanced by gender and accent, and totaling 70.8 hours ($\approx$4,249 minutes) of speech with 11,200 questions. Using four complementary metrics (accuracy, entropy, APES, and Fleiss' $κ$), we evaluate nine representative models under linguistic (language and accent), demographic (gender), and structural (option order) perturbations. Our findings reveal that MLLMs are relatively robust to demographic factors but highly sensitive to language and option order, suggesting that speech can amplify existing structural biases. Moreover, architectural design and reasoning strategy substantially affect robustness across languages. Overall, this study establishes a unified framework for assessing fairness and robustness in speech-integrated LLMs, bridging the gap between text- and speech-based evaluation. The resources can be found at https://github.com/ntunlplab/BiasInEar.
- Open Access
Beyond Known Facts: Generating Unseen Temporal Knowledge to Address Data Contamination in LLM Evaluation
The automatic extraction of information is important for populating large web knowledge bases such as Wikidata. The temporal version of that task, temporal knowledge graph extraction (TKGE), involves extracting temporally grounded facts from text, represented as semantic quadruples (subject, relation, object, timestamp). Many recent systems take advantage of large language models (LLMs), which are becoming a new cornerstone of the web due to their performance on many tasks across the natural language processing (NLP) field. Despite the importance of TKGE, existing datasets for training and evaluation remain scarce, and contamination of evaluation data is an unaddressed issue, potentially inflating LLMs' perceived performance due to overlaps between training and evaluation sets. To mitigate these challenges, we propose a novel synthetic evaluation dataset constructed from predicted future, previously unseen temporal facts, thereby eliminating contamination and enabling robust and unbiased benchmarking. Our dataset creation involves a two-step approach: (1) Temporal Knowledge Graph Forecasting (TKGF) generates plausible future quadruples, which are subsequently filtered to adhere to the original knowledge base schema; (2) LLMs perform quadruple-to-text generation, creating semantically aligned textual descriptions. We benchmark Extract, Define and Canonicalize (EDC), a state-of-the-art LLM-based extraction framework, demonstrating that LLM performance decreases when evaluated on our dataset compared to a dataset of known facts. We publicly release our dataset consisting of 4.2K future quadruples and corresponding textual descriptions, along with the generation methodology, enabling continuous creation of unlimited future temporal datasets to serve as long-term, contamination-free benchmarks for TKGE.
Open AccessBeyond Known Facts: Generating Unseen Temporal Knowledge to Address Data Contamination in LLM Evaluation
The automatic extraction of information is important for populating large web knowledge bases such as Wikidata. The temporal version of that task, temporal knowledge graph extraction (TKGE), involves extracting temporally grounded facts from text, represented as semantic quadruples (subject, relation, object, timestamp). Many recent systems take advantage of large language models (LLMs), which are becoming a new cornerstone of the web due to their performance on many tasks across the natural language processing (NLP) field. Despite the importance of TKGE, existing datasets for training and evaluation remain scarce, and contamination of evaluation data is an unaddressed issue, potentially inflating LLMs' perceived performance due to overlaps between training and evaluation sets. To mitigate these challenges, we propose a novel synthetic evaluation dataset constructed from predicted future, previously unseen temporal facts, thereby eliminating contamination and enabling robust and unbiased benchmarking. Our dataset creation involves a two-step approach: (1) Temporal Knowledge Graph Forecasting (TKGF) generates plausible future quadruples, which are subsequently filtered to adhere to the original knowledge base schema; (2) LLMs perform quadruple-to-text generation, creating semantically aligned textual descriptions. We benchmark Extract, Define and Canonicalize (EDC), a state-of-the-art LLM-based extraction framework, demonstrating that LLM performance decreases when evaluated on our dataset compared to a dataset of known facts. We publicly release our dataset consisting of 4.2K future quadruples and corresponding textual descriptions, along with the generation methodology, enabling continuous creation of unlimited future temporal datasets to serve as long-term, contamination-free benchmarks for TKGE.
- Open Access
- 10 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 4659-4664
VQA-Induct: Instruction Induction for Visual Question Answering
- 10 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 4659-4664
Open AccessVQA-Induct: Instruction Induction for Visual Question Answering
- Open Access
- 10 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 5439-5443
AR 2 : Adversarial Reinforcement Learning for Abstract Reasoning in Large Language Models
- 10 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 5439-5443
Open AccessAR 2 : Adversarial Reinforcement Learning for Abstract Reasoning in Large Language Models
- Open Access
- 10 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 5464-5468
Robust Handwritten Text Recognition via Multi-Source Adversarial Domain Adaptation for Low-Resource Scripts
- 10 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 5464-5468
Open AccessRobust Handwritten Text Recognition via Multi-Source Adversarial Domain Adaptation for Low-Resource Scripts
- Open Access
Visual Lifelog Retrieval through Captioning-Enhanced Interpretation
People often struggle to remember specific details of past experiences, which can lead to the need to revisit these memories. Consequently, lifelog retrieval has emerged as a crucial application. Various studies have explored methods to facilitate rapid access to personal lifelogs for memory recall assistance. In this paper, we propose a Captioning-Integrated Visual Lifelog (CIVIL) Retrieval System for extracting specific images from a user's visual lifelog based on textual queries. Unlike traditional embedding-based methods, our system first generates captions for visual lifelogs and then utilizes a text embedding model to project both the captions and user queries into a shared vector space. Visual lifelogs, captured through wearable cameras, provide a first-person viewpoint, necessitating the interpretation of the activities of the individual behind the camera rather than merely describing the scene. To address this, we introduce three distinct approaches: the single caption method, the collective caption method, and the merged caption method, each designed to interpret the life experiences of lifeloggers. Experimental results show that our method effectively describes first-person visual images, enhancing the outcomes of lifelog retrieval. Furthermore, we construct a textual dataset that converts visual lifelogs into captions, thereby reconstructing personal life experiences.
Open AccessVisual Lifelog Retrieval through Captioning-Enhanced Interpretation
People often struggle to remember specific details of past experiences, which can lead to the need to revisit these memories. Consequently, lifelog retrieval has emerged as a crucial application. Various studies have explored methods to facilitate rapid access to personal lifelogs for memory recall assistance. In this paper, we propose a Captioning-Integrated Visual Lifelog (CIVIL) Retrieval System for extracting specific images from a user's visual lifelog based on textual queries. Unlike traditional embedding-based methods, our system first generates captions for visual lifelogs and then utilizes a text embedding model to project both the captions and user queries into a shared vector space. Visual lifelogs, captured through wearable cameras, provide a first-person viewpoint, necessitating the interpretation of the activities of the individual behind the camera rather than merely describing the scene. To address this, we introduce three distinct approaches: the single caption method, the collective caption method, and the merged caption method, each designed to interpret the life experiences of lifeloggers. Experimental results show that our method effectively describes first-person visual images, enhancing the outcomes of lifelog retrieval. Furthermore, we construct a textual dataset that converts visual lifelogs into captions, thereby reconstructing personal life experiences.
- Open Access
Do Before You Judge: Self-Reference as a Pathway to Better LLM Evaluation
LLM-as-Judge frameworks are increasingly popular for AI evaluation, yet research findings on the relationship between models' generation and judgment abilities remain inconsistent. We investigate this relationship through systematic dataset- and instance-level analyses across 11 models and 21 diverse tasks. Despite both capabilities relying on the same underlying knowledge, our analyses reveal they are only weakly correlated, primarily due to LLMs' sensitivity to the responses being judged. To address this, we propose a self-reference-guided evaluation strategy that leverages a model's own answers as references. This approach significantly strengthens the correlation between generation and judgment abilities, offering a practical path to align these skills and providing a reliable proxy for model selection in evaluation tasks.
Open AccessDo Before You Judge: Self-Reference as a Pathway to Better LLM Evaluation
LLM-as-Judge frameworks are increasingly popular for AI evaluation, yet research findings on the relationship between models' generation and judgment abilities remain inconsistent. We investigate this relationship through systematic dataset- and instance-level analyses across 11 models and 21 diverse tasks. Despite both capabilities relying on the same underlying knowledge, our analyses reveal they are only weakly correlated, primarily due to LLMs' sensitivity to the responses being judged. To address this, we propose a self-reference-guided evaluation strategy that leverages a model's own answers as references. This approach significantly strengthens the correlation between generation and judgment abilities, offering a practical path to align these skills and providing a reliable proxy for model selection in evaluation tasks.
- Open Access
Diagnosing Model Editing via Knowledge Spectrum
Model editing, the process of efficiently modifying factual knowledge in pre-trained language models, is critical for maintaining their accuracy and relevance. However, existing editing methods often introduce unintended side effects, degrading model performance in unpredictable ways. While much research has focused on improving editing algorithms, the role of the target knowledge's intrinsic properties remains a significant, underexplored factor. This paper addresses this gap by first proposing the ``Knowledge Spectrum,'' a systematic framework for categorizing knowledge based on its real-world popularity, the model's pre-edit familiarity, and the linguistic structure of the eliciting question. Our empirical analysis reveals that these characteristics are strong predictors of editing success and stability. Informed by these findings, we introduce the ``Knowledge-Diagnostic Framework,'' an adaptive strategy that tailors editing intensity to the diagnosed difficulty of a knowledge item. We demonstrate that this framework significantly improves success rates for challenging edits while optimizing computational resources. Our work provides a more comprehensive understanding of the factors governing model editing.
Open AccessDiagnosing Model Editing via Knowledge Spectrum
Model editing, the process of efficiently modifying factual knowledge in pre-trained language models, is critical for maintaining their accuracy and relevance. However, existing editing methods often introduce unintended side effects, degrading model performance in unpredictable ways. While much research has focused on improving editing algorithms, the role of the target knowledge's intrinsic properties remains a significant, underexplored factor. This paper addresses this gap by first proposing the ``Knowledge Spectrum,'' a systematic framework for categorizing knowledge based on its real-world popularity, the model's pre-edit familiarity, and the linguistic structure of the eliciting question. Our empirical analysis reveals that these characteristics are strong predictors of editing success and stability. Informed by these findings, we introduce the ``Knowledge-Diagnostic Framework,'' an adaptive strategy that tailors editing intensity to the diagnosed difficulty of a knowledge item. We demonstrate that this framework significantly improves success rates for challenging edits while optimizing computational resources. Our work provides a more comprehensive understanding of the factors governing model editing.