arXiv subjects

Xiaozheng Wang

Publications and source records attributed to Xiaozheng Wang.

5 recordsLinked to original sources

arXiv2026

Generating Synthetic Behavioral Populations from XR Motion

Large-scale behavioral datasets are becoming increasingly important for machine learning, personalization, and behavioral modeling in extended reality (XR). However, collecting XR motion data from hundreds or thousands of participants remains expensive, time-consuming, and difficult to reproduce across research groups. As a result, many XR studies continue to rely on relatively small datasets that limit the scale and diversity of behavioral evaluation. To address this limitation, we investigate synthetic behavioral populations as a complementary approach to traditional XR data collection. We present an interpolation-based motion synthesis pipeline that combines dynamic time warping (DTW) with trajectory interpolation to generate synthetic behavioral trajectories from existing XR datasets while preserving task structure and incorporating motion characteristics from contributing participants. Using the publicly available FAST VR assembly dataset, we generated and openly released 100 synthetic behavioral trajectories. We evaluated the synthesized trajectories through motion-based user identification. Hybrid datasets containing both real and synthesized trajectories achieved performance comparable to similarly sized real-only datasets while maintaining low confusion between synthesized trajectories and their contributing participants. Rather than serving as conventional data augmentation, the proposed approach generates distinguishable behavioral trajectories that expand XR behavioral populations for larger-scale behavioral modeling and machine learning evaluation. Our findings demonstrate that synthetic behavioral populations provide a promising approach to expanding XR behavioral datasets and supporting future data-driven immersive systems.

cs.HC↗

arXiv2026

The Capturing and Logging Ecological Virtual Experiences and Reality (CLEVER) - Job Simulator Dataset

Virtual reality (VR) motion tracking and interaction data has become increasingly recognized as valuable for machine learning experiments for a variety of purposes, including predicting user identities, predicting user attributes like gender and age, predicting retention and learning, and more. However, there exist a limited number of publicly accessible VR motion datasets. In this paper, we present a new open-source dataset of 95 participants playing the SteamVR game Job Simulator. Additionally, we review existing datasets, detail our study procedure, describe our data collection process, list attributes of our dataset, and suggest future work, impact, and applications.

cs.HC↗

arXiv2025

Broad Critic Deep Actor Reinforcement Learning for Continuous Control

In the domain of continuous control, deep reinforcement learning (DRL) demonstrates promising results. However, the dependence of DRL on deep neural networks (DNNs) results in the demand for extensive data and increased computational cost. To address this issue, a novel hybrid actor-critic reinforcement learning (RL) framework is introduced. The proposed framework integrates the broad learning system (BLS) with DNN, aiming to merge the strengths of both distinct architectural paradigms. Specifically, the critic network employs BLS for rapid value estimation via ridge regression, while the actor network retains the DNN structure to optimize policy gradients. This hybrid design is generalizable and can enhance existing actor-critic algorithms. To demonstrate its versatility, the proposed framework is integrated into three widely used actor-critic algorithms -- deep deterministic policy gradient (DDPG), soft actor-critic (SAC), and twin delayed DDPG (TD3), resulting in BLS-augmented variants. Experimental results reveal that all BLS-enhanced versions surpass their original counterparts in terms of training efficiency and accuracy. These improvements highlight the suitability of the proposed framework for real-time control scenarios, where computational efficiency and rapid adaptation are critical.

cs.LG↗

arXiv2025

Users' Mental Models of Generative AI Chatbot Ecosystems

The capability of GenAI-based chatbots, such as ChatGPT and Gemini, has expanded quickly in recent years, turning them into GenAI Chatbot Ecosystems. Yet, users' understanding of how such ecosystems work remains unknown. In this paper, we investigate users' mental models of how GenAI Chatbot Ecosystems work. This is an important question because users' mental models guide their behaviors, including making decisions that impact their privacy. Through 21 semi-structured interviews, we uncovered users' four mental models towards first-party (e.g., Google Gemini) and third-party (e.g., ChatGPT) GenAI Chatbot Ecosystems. These mental models centered around the role of the chatbot in the entire ecosystem. We further found that participants held a more consistent and simpler mental model towards third-party ecosystems than the first-party ones, resulting in higher trust and fewer concerns towards the third-party ecosystems. We discuss the design and policy implications based on our results.

cs.HC↗

arXiv2023

Exploration of visual prompt in Grounded pre-trained open-set detection

Text prompts are crucial for generalizing pre-trained open-set object detection models to new categories. However, current methods for text prompts are limited as they require manual feedback when generalizing to new categories, which restricts their ability to model complex scenes, often leading to incorrect detection results. To address this limitation, we propose a novel visual prompt method that learns new category knowledge from a few labeled images, which generalizes the pre-trained detection model to the new category. To allow visual prompts to represent new categories adequately, we propose a statistical-based prompt construction module that is not limited by predefined vocabulary lengths, thus allowing more vectors to be used when representing categories. We further utilize the category dictionaries in the pre-training dataset to design task-specific similarity dictionaries, which make visual prompts more discriminative. We evaluate the method on the ODinW dataset and show that it outperforms existing prompt learning methods and performs more consistently in combinatorial inference.

cs.CV↗