Ting-Hao Huang
YOU?
Author Swipe
View article: Hashtag Re-Appropriation for Audience Control on Recommendation-Driven Social Media Xiaohongshu (rednote)
Hashtag Re-Appropriation for Audience Control on Recommendation-Driven Social Media Xiaohongshu (rednote) Open
Algorithms have played a central role in personalized recommendations on social media. However, they also present significant obstacles for content creators trying to predict and manage their audience reach. This issue is particularly chal…
View article: Prompting in the Dark: Assessing Human Performance in Prompt Engineering for Data Labeling When Gold Labels Are Absent
Prompting in the Dark: Assessing Human Performance in Prompt Engineering for Data Labeling When Gold Labels Are Absent Open
Millions of users prompt large language models (LLMs) for various tasks, but how good are people at prompt engineering? Do users actually get closer to their desired outcome over multiple iterations of their prompts? These questions are cr…
View article: From Selection to Generation: A Survey of LLM-based Active Learning
From Selection to Generation: A Survey of LLM-based Active Learning Open
Active Learning (AL) has been a powerful paradigm for improving model efficiency and performance by selecting the most informative data points for labeling and training. In recent active learning frameworks, Large Language Models (LLMs) ha…
View article: Using Contextually Aligned Online Reviews to Measure LLMs' Performance Disparities Across Language Varieties
Using Contextually Aligned Online Reviews to Measure LLMs' Performance Disparities Across Language Varieties Open
A language can have different varieties. These varieties can affect the performance of natural language processing (NLP) models, including large language models (LLMs), which are often trained on data from widely spoken varieties. This pap…
View article: Do Large Multimodal Models Solve Caption Generation for Scientific Figures? Lessons Learned from SciCap Challenge 2023
Do Large Multimodal Models Solve Caption Generation for Scientific Figures? Lessons Learned from SciCap Challenge 2023 Open
Since the SciCap datasets launch in 2021, the research community has made significant progress in generating captions for scientific figures in scholarly articles. In 2023, the first SciCap Challenge took place, inviting global teams to us…
View article: Multi-LLM Collaborative Caption Generation in Scientific Documents
Multi-LLM Collaborative Caption Generation in Scientific Documents Open
Scientific figure captioning is a complex task that requires generating contextually appropriate descriptions of visual content. However, existing methods often fall short by utilizing incomplete information, treating the task solely as ei…
View article: Front Matter
Front Matter Open
View article: Understanding Writing Assistants for Scientific Figure Captions: A Thematic Analysis
Understanding Writing Assistants for Scientific Figure Captions: A Thematic Analysis Open
View article: From Noise to Nuance: Enriching Subjective Data Annotation through Qualitative Analysis
From Noise to Nuance: Enriching Subjective Data Annotation through Qualitative Analysis Open
View article: LaMP-Cap: Personalized Figure Caption Generation With Multimodal Figure Profiles
LaMP-Cap: Personalized Figure Caption Generation With Multimodal Figure Profiles Open
View article: From Selection to Generation: A Survey of LLM-based Active Learning
From Selection to Generation: A Survey of LLM-based Active Learning Open
View article: Using Contextually Aligned Online Reviews to Measure LLMs’ Performance Disparities Across Language Varieties
Using Contextually Aligned Online Reviews to Measure LLMs’ Performance Disparities Across Language Varieties Open
View article: Effectiveness of pre-Omicron COVID-19 mRNA vaccines against hospitalizations in infection-naïve children: a case-control study
Effectiveness of pre-Omicron COVID-19 mRNA vaccines against hospitalizations in infection-naïve children: a case-control study Open
This case–control study spanned from July 2022 to September 2023, focusing on the pre-Omicron COVID-19 mRNA vaccine effectiveness against the Omicron variant in children without prior SARS-CoV-2 exposure. We reported that pre-Omicron COVID…
View article: DetailSemNet: Elevating Signature Verification Through Detail-Semantic Integration
DetailSemNet: Elevating Signature Verification Through Detail-Semantic Integration Open
View article: CoCoLoFa: A Dataset of News Comments with Common Logical Fallacies Written by LLM-Assisted Crowds
CoCoLoFa: A Dataset of News Comments with Common Logical Fallacies Written by LLM-Assisted Crowds Open
Detecting logical fallacies in texts can help users spot argument flaws, but automating this detection is not easy. Manually annotating fallacies in large-scale, real-world text data to create datasets for developing and validating detecti…
View article: What Color Scheme is More Effective in Assisting Readers to Locate Information in a Color-Coded Article?
What Color Scheme is More Effective in Assisting Readers to Locate Information in a Color-Coded Article? Open
Color coding, a technique assigning specific colors to cluster information types, has proven advantages in aiding human cognitive activities, especially reading and comprehension. The rise of Large Language Models (LLMs) has streamlined do…
View article: Generating Educational Materials with Different Levels of Readability using LLMs
Generating Educational Materials with Different Levels of Readability using LLMs Open
This study introduces the leveled-text generation task, aiming to rewrite educational materials to specific readability levels while preserving meaning. We assess the capability of GPT-3.5, LLaMA-2 70B, and Mixtral 8x7B, to generate conten…
View article: On Replacing Humans with Large Language Models in Voice-Based Human-in-the-Loop Systems
On Replacing Humans with Large Language Models in Voice-Based Human-in-the-Loop Systems Open
It is easy to assume that Large Language Models (LLMs) will seamlessly take over applications, especially those that are largely automated. In the case of conversational voice assistants, commercial systems have been widely deployed and us…
View article: Dark Sides: Envisioning, Understanding, and Preventing Harmful Effects of Writing Assistants - The Third Workshop on Intelligent and Interactive Writing Assistants
Dark Sides: Envisioning, Understanding, and Preventing Harmful Effects of Writing Assistants - The Third Workshop on Intelligent and Interactive Writing Assistants Open
Writing assistants are becoming increasingly sophisticated and ubiquitous, fueled by advances in artificial intelligence, particularly large language models. As new use cases and models emerge, we expect the adoption rate to accelerate. Th…
View article: If in a Crowdsourced Data Annotation Pipeline, a GPT-4
If in a Crowdsourced Data Annotation Pipeline, a GPT-4 Open
Recent studies indicated GPT-4 outperforms online crowd workers in data\nlabeling accuracy, notably workers from Amazon Mechanical Turk (MTurk).\nHowever, these studies were criticized for deviating from standard\ncrowdsourcing practices a…
View article: How Does Conversation Length Impact User’s Satisfaction? A Case Study of Length-Controlled Conversations with LLM-Powered Chatbots
How Does Conversation Length Impact User’s Satisfaction? A Case Study of Length-Controlled Conversations with LLM-Powered Chatbots Open
Users can discuss a wide range of topics with large language models (LLMs), but they do not always prefer solving problems or getting information through lengthy conversations. This raises an intriguing HCI question: How does instructing L…
View article: SciCapenter: Supporting Caption Composition for Scientific Figures with Machine-Generated Captions and Ratings
SciCapenter: Supporting Caption Composition for Scientific Figures with Machine-Generated Captions and Ratings Open
Crafting effective captions for figures is important. Readers heavily depend\non these captions to grasp the figure's message. However, despite a\nwell-developed set of AI technologies for figures and captions, these have\nrarely been test…
View article: How Does Conversation Length Impact User's Satisfaction? A Case Study of Length-Controlled Conversations with LLM-Powered Chatbots
How Does Conversation Length Impact User's Satisfaction? A Case Study of Length-Controlled Conversations with LLM-Powered Chatbots Open
Users can discuss a wide range of topics with large language models (LLMs), but they do not always prefer solving problems or getting information through lengthy conversations. This raises an intriguing HCI question: How does instructing L…
View article: Inspo: Writing with Crowds Alongside AI
Inspo: Writing with Crowds Alongside AI Open
The use of artificial intelligence (AI) to support creative writing has bloomed in recent years. However, it is less well understood how AI compares to on-demand human support. We explored how writers interact with both AI and crowd worker…
View article: Does Human Collaboration Enhance the Accuracy of Identifying LLM-Generated Deepfake Texts?
Does Human Collaboration Enhance the Accuracy of Identifying LLM-Generated Deepfake Texts? Open
Advances in Large Language Models (e.g., GPT-4, LLaMA) have improved the generation of coherent sentences resembling human writing on a large scale, resulting in the creation of so-called deepfake texts. However, this progress poses securi…
View article: Location-Aware Visual Question Generation with Lightweight Models
Location-Aware Visual Question Generation with Lightweight Models Open
This work introduces a novel task, location-aware visual question generation (LocaVQG), which aims to generate engaging questions from data relevant to a particular geographical location. Specifically, we represent such location-aware info…
View article: GPT-4 as an Effective Zero-Shot Evaluator for Scientific Figure Captions
GPT-4 as an Effective Zero-Shot Evaluator for Scientific Figure Captions Open
There is growing interest in systems that generate captions for scientific figures. However, assessing these systems output poses a significant challenge. Human evaluation requires academic expertise and is costly, while automatic evaluati…
View article: Good Data, Large Data, or No Data? Comparing Three Approaches in Developing Research Aspect Classifiers for Biomedical Papers
Good Data, Large Data, or No Data? Comparing Three Approaches in Developing Research Aspect Classifiers for Biomedical Papers Open
The rapid growth of scientific publications, particularly during the COVID-19 pandemic, emphasizes the need for tools to help researchers efficiently comprehend the latest advancements. One essential part of understanding scientific litera…
View article: ConvXAI: Delivering Heterogeneous AI Explanations via Conversations to Support Human-AI Scientific Writing
ConvXAI: Delivering Heterogeneous AI Explanations via Conversations to Support Human-AI Scientific Writing Open
Despite a surge collection of XAI methods, users still struggle to obtain required AI explanations. Previous research suggests chatbots as dynamic solutions, but the effective design of conversational XAI agents for practical human needs r…
View article: Does Human Collaboration Enhance the Accuracy of Identifying LLM-Generated Deepfake Texts?
Does Human Collaboration Enhance the Accuracy of Identifying LLM-Generated Deepfake Texts? Open
Advances in Large Language Models (e.g., GPT-4, LLaMA) have improved the generation of coherent sentences resembling human writing on a large scale, resulting in the creation of so-called deepfake texts. However, this progress poses securi…