arXiv subjects

Haein Kong

Publications and source records attributed to Haein Kong.

7 recordsLinked to original sources

arXiv2026

Child Safety in Generative AI: An Expert-Guided and Incident-Grounded Evaluation Framework

As generative AI is increasingly used by children and adolescents, there is a growing need for risk evaluation frameworks that account for child-specific harms. However, most existing safety evaluation frameworks focus on general user populations, often overlooking risks unique to younger users. To address this gap, we propose an evaluation framework that integrates expert-guided risk factors with real-world AI incident data for child safety. The framework identifies hazard categories from expert guidelines and AI incident databases and uses this information to construct a synthetic test set for model evaluation. Particularly, we apply the framework to the education domain and evaluate three Llama Guard models on their ability to detect unsafe user prompts. Our results show that current Llama Guard models struggle to identify education-related unsafe user prompts. We conclude by discussing how future work can extend the evaluation to additional risk categories and incorporate domain experts throughout the evaluation pipeline.

cs.HC↗

arXiv2025

Judging from Support-set: A New Way to Utilize Few-Shot Segmentation for Segmentation Refinement Process

Segmentation refinement aims to enhance the initial coarse masks generated by segmentation algorithms. The refined masks are expected to capture more details and better contours of the target objects. Research on segmentation refinement has developed as a response to the need for high-quality image segmentations. However, to our knowledge, no method has been developed that can determine the success of segmentation refinement. Such a method could ensure the reliability of segmentation in applications where the outcome of the segmentation is important and fosters innovation in image processing technologies. To address this research gap, we propose Judging From Support-set (JFS), a method to judge the success of segmentation refinement leveraging an off-the-shelf few-shot segmentation (FSS) model. The traditional goal of the problem in FSS is to find a target object in a query image utilizing target information given by a support set. However, we propose a novel application of the FSS model in our evaluation pipeline for segmentation refinement methods. Given a coarse mask as input, segmentation refinement methods produce a refined mask; these two masks become new support masks for the FSS model. The existing support mask then serves as the test set for the FSS model to evaluate the quality of the refined segmentation by the segmentation refinement methods. We demonstrate the effectiveness of our proposed JFS framework by evaluating the SAM Enhanced Pseudo-Labels (SEPL) using SegGPT as the choice of FSS model on the PASCAL dataset. The results showed that JFS has the potential to determine whether the segmentation refinement process is successful.

cs.CV↗

arXiv2025

Persuasion and Safety in the Era of Generative AI

As large language models (LLMs) achieve advanced persuasive capabilities, concerns about their potential risks have grown. The EU AI Act prohibits AI systems that use manipulative or deceptive techniques to undermine informed decision-making, highlighting the need to distinguish between rational persuasion, which engages reason, and manipulation, which exploits cognitive biases. My dissertation addresses the lack of empirical studies in this area by developing a taxonomy of persuasive techniques, creating a human-annotated dataset, and evaluating LLMs' ability to distinguish between these methods. This work contributes to AI safety by providing resources to mitigate the risks of persuasive AI and fostering discussions on ethical persuasion in the age of generative AI.

cs.CY↗

arXiv2025

Examining Racial Stereotypes in YouTube Autocomplete Suggestions

Autocomplete is a popular search feature that predicts queries based on user input and guides users to a set of potentially relevant suggestions. In this study, we examine what YouTube autocompletes suggest to users seeking information about race on the platform. Specifically, we perform an algorithm output audit of autocomplete suggestions for input queries about four racial groups and examine the stereotypes they embody. Using critical discourse analysis, we identify five major sociocultural contexts in which racial information appears -Appearance, Ability, Culture, Social Equity, and Manner. We found that the participatory nature of YouTube produces a multifaceted representation of race-related content in its search outputs, characterized by enduring historical biases, aggregated discrimination, and interracial tensions, while simultaneously depicting minority resistance and aspirations of a post-racial society. We call for innovations in content moderation policy design and enforcement to address existing racial harms in YouTube search outputs.

cs.CY↗

arXiv2025

When LLM Therapists Become Salespeople: Evaluating Large Language Models for Ethical Motivational Interviewing

Large language models (LLMs) have been actively applied in the mental health field. Recent research shows the promise of LLMs in applying psychotherapy, especially motivational interviewing (MI). However, there is a lack of studies investigating how language models understand MI ethics. Given the risks that malicious actors can use language models to apply MI for unethical purposes, it is important to evaluate their capability of differentiating ethical and unethical MI practices. Thus, this study investigates the ethical awareness of LLMs in MI with multiple experiments. Our findings show that LLMs have a moderate to strong level of knowledge in MI. However, their ethical standards are not aligned with the MI spirit, as they generated unethical responses and performed poorly in detecting unethical responses. We proposed a Chain-of-Ethic prompt to mitigate those risks and improve safety. Finally, our proposed strategy effectively improved ethical MI response generation and detection performance. These findings highlight the need for safety evaluations and guidelines for building ethical LLM-powered psychotherapy.

cs.CL↗

arXiv2024

Gender Bias in LLM-generated Interview Responses

LLMs have emerged as a promising tool for assisting individuals in diverse text-generation tasks, including job-related texts. However, LLM-generated answers have been increasingly found to exhibit gender bias. This study evaluates three LLMs (GPT-3.5, GPT-4, Claude) to conduct a multifaceted audit of LLM-generated interview responses across models, question types, and jobs, and their alignment with two gender stereotypes. Our findings reveal that gender bias is consistent, and closely aligned with gender stereotypes and the dominance of jobs. Overall, this study contributes to the systematic examination of gender bias in LLM-generated interview responses, highlighting the need for a mindful approach to mitigate such biases in related applications.

cs.CL↗

arXiv2024

FCC: Fully Connected Correlation for One-Shot Segmentation

Few-shot segmentation (FSS) aims to segment the target object in a query image using only a small set of support images and masks. Therefore, having strong prior information for the target object using the support set is essential for guiding the initial training of FSS, which leads to the success of few-shot segmentation in challenging cases, such as when the target object shows considerable variation in appearance, texture, or scale across the support and query images. Previous methods have tried to obtain prior information by creating correlation maps from pixel-level correlation on final-layer or same-layer features. However, we found these approaches can offer limited and partial information when advanced models like Vision Transformers are used as the backbone. Vision Transformer encoders have a multi-layer structure with identical shapes in their intermediate layers. Leveraging the feature comparison from all layers in the encoder can enhance the performance of few-shot segmentation. We introduce FCC (Fully Connected Correlation) to integrate pixel-level correlations between support and query features, capturing associations that reveal target-specific patterns and correspondences in both same-layers and cross-layers. FCC captures previously inaccessible target information, effectively addressing the limitations of support mask. Our approach consistently demonstrates state-of-the-art performance on PASCAL, COCO, and domain shift tests. We conducted an ablation study and cross-layer correlation analysis to validate FCC's core methodology. These findings reveal the effectiveness of FCC in enhancing prior information and overall model performance.

cs.CV↗