Chang Xiao
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Chang Xiao has a total of 34 co-authors. Here are the top co-authors by number of shared publications:
- Eunyee Koh
Adobe Research, San Jose, California, USA
5 shared publications - Alexa Siu
Adobe Research, San Jose, California, USA
3 shared publications - Mustafa Doga Dogan
Adobe Research, Basel, Switzerland
3 shared publications - Gromit Yeuk-Yin Chan
Adobe Research, San Jose, California, USA
2 shared publications - Jane Hoffswell
Adobe Research, Seattle, Washington, USA
2 shared publications - Ryan Anthony Rossi
Adobe Research, San Jose, USA
2 shared publications
Latest Publications
- Open Access
- 2 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-3
STRAIGHT: Toward Consistent Archery Form via Video Pose Analysis and Language-Model-Based Coaching Feedback
- 2 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-3
Open AccessSTRAIGHT: Toward Consistent Archery Form via Video Pose Analysis and Language-Model-Based Coaching Feedback
- Open Access
- 2 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-13
TangiMap: Context-Aware Generative Function Mapping for Tangible Input Devices
While tangible input devices with physical controls such as knobs and sliders enable efficient interaction, they often lack the flexibility to adapt to dynamic contexts, limiting their use to predefined applications with fixed functionalities. In this paper, we introduce TangiMap, a proof-of-concept system that bridges tangible input devices with dynamic software contexts by generating adaptive function mappings for hardware controls. To achieve this, TangiMap analyzes the user’s declared hardware specifications, the available controls, and the target software context, and uses large language models (LLMs) to assign suitable functions to each control element. To support diverse contexts, TangiMap presents multiple strategies such as API queries and screen parsing. We demonstrate the system through four scenarios: interactive adjustment of natural language parameters in 3D modeling, live music performance, task-aware mapping of on-screen controls in image editing, and dynamic puppeteering in animation.
- 2 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-13
Open AccessTangiMap: Context-Aware Generative Function Mapping for Tangible Input Devices
While tangible input devices with physical controls such as knobs and sliders enable efficient interaction, they often lack the flexibility to adapt to dynamic contexts, limiting their use to predefined applications with fixed functionalities. In this paper, we introduce TangiMap, a proof-of-concept system that bridges tangible input devices with dynamic software contexts by generating adaptive function mappings for hardware controls. To achieve this, TangiMap analyzes the user’s declared hardware specifications, the available controls, and the target software context, and uses large language models (LLMs) to assign suitable functions to each control element. To support diverse contexts, TangiMap presents multiple strategies such as API queries and screen parsing. We demonstrate the system through four scenarios: interactive adjustment of natural language parameters in 3D modeling, live music performance, task-aware mapping of on-screen controls in image editing, and dynamic puppeteering in animation.
- Open Access
- 5 October 2025
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Human-Computer Interaction
- Vol. 9 (6) , 1097-1123
LLMs May Not Be Human-Level Players, But They Can Be Testers: Measuring Game Difficulty with LLM Agents
Recent advances in Large Language Models (LLMs) have demonstrated their potential as autonomous agents across various tasks. One emerging application is the use of LLMs in playing games. In this work, we explore a practical problem for the gaming industry: Can LLMs be used to measure game difficulty? We evaluate the feasibility of using LLM agents to test game difficulty, focusing on two widely played games: Wordle and Slay the Spire . Our results reveal an interesting finding: although LLMs may not perform as well as the average human player, their performance, when guided by simple, generic prompting techniques, shows a statistically significant and strong correlation with difficulty indicated by human players. This suggests that LLMs could potentially serve as human-like agents for measuring game difficulty during the development process, as their assessments may align closely with those of human players. Based on our experiments, we also propose general principles and guidelines for integrating LLMs into the game testing workflow.
- 5 October 2025
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Human-Computer Interaction
- Vol. 9 (6) , 1097-1123
Open AccessLLMs May Not Be Human-Level Players, But They Can Be Testers: Measuring Game Difficulty with LLM Agents
Recent advances in Large Language Models (LLMs) have demonstrated their potential as autonomous agents across various tasks. One emerging application is the use of LLMs in playing games. In this work, we explore a practical problem for the gaming industry: Can LLMs be used to measure game difficulty? We evaluate the feasibility of using LLM agents to test game difficulty, focusing on two widely played games: Wordle and Slay the Spire . Our results reveal an interesting finding: although LLMs may not perform as well as the average human player, their performance, when guided by simple, generic prompting techniques, shows a statistically significant and strong correlation with difficulty indicated by human players. This suggests that LLMs could potentially serve as human-like agents for measuring game difficulty during the development process, as their assessments may align closely with those of human players. Based on our experiments, we also propose general principles and guidelines for integrating LLMs into the game testing workflow.
- Open Access
- 27 September 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-13
Streaming, Fast and Slow: Cognitive Load-Aware Streaming for Efficient LLM Serving
- 27 September 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-13
Open AccessStreaming, Fast and Slow: Cognitive Load-Aware Streaming for Efficient LLM Serving
- Open Access
- 25 April 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-18
Imprinto: Enhancing Infrared Inkjet Watermarking for Human and Machine Perception
- 25 April 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-18
Open AccessImprinto: Enhancing Infrared Inkjet Watermarking for Human and Machine Perception
- Open Access
- 4 March 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-12
ReactFold: Towards Camera-based Tangible Interaction on Passive Paper Artifacts
- 4 March 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-12
Open AccessReactFold: Towards Camera-based Tangible Interaction on Passive Paper Artifacts
- Open Access
- 19 November 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Graphics
- Vol. 43 (6) , 1-12
Evaluating Visual Perception of Object Motion in Dynamic Environments
Precisely understanding how objects move in 3D is essential for broad scenarios such as video editing, gaming, driving, and athletics. With screen-displayed computer graphics content, users only perceive limited cues to judge the object motion from the on-screen optical flow. Conventionally, visual perception is studied with stationary settings and singular objects. However, in practical applications, we---the observer---also move within complex scenes. Therefore, we must extract object motion from a combined optical flow displayed on screen, which can often lead to mis-estimations due to perceptual ambiguities. We measure and model observers' perceptual accuracy of object motions in dynamic 3D environments, a universal but under-investigated scenario in computer graphics applications. We design and employ a crowdsourcing-based psychophysical study, quantifying the relationships among patterns of scene dynamics and content, and the resulting perceptual judgments of object motion direction. The acquired psychophysical data underpins a model for generalized conditions. We then demonstrate the model's guidance ability to significantly enhance users' understanding of task object motion in gaming and animation design. With applications in measuring and compensating for object motion errors in video and rendering, we hope the research establishes a new frontier for understanding and mitigating perceptual errors caused by the gap between screen-displayed graphics and the physical world.
- 19 November 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Graphics
- Vol. 43 (6) , 1-12
Open AccessEvaluating Visual Perception of Object Motion in Dynamic Environments
Precisely understanding how objects move in 3D is essential for broad scenarios such as video editing, gaming, driving, and athletics. With screen-displayed computer graphics content, users only perceive limited cues to judge the object motion from the on-screen optical flow. Conventionally, visual perception is studied with stationary settings and singular objects. However, in practical applications, we---the observer---also move within complex scenes. Therefore, we must extract object motion from a combined optical flow displayed on screen, which can often lead to mis-estimations due to perceptual ambiguities. We measure and model observers' perceptual accuracy of object motions in dynamic 3D environments, a universal but under-investigated scenario in computer graphics applications. We design and employ a crowdsourcing-based psychophysical study, quantifying the relationships among patterns of scene dynamics and content, and the resulting perceptual judgments of object motion direction. The acquired psychophysical data underpins a model for generalized conditions. We then demonstrate the model's guidance ability to significantly enhance users' understanding of task object motion in gaming and animation design. With applications in measuring and compensating for object motion errors in video and rendering, we hope the research establishes a new frontier for understanding and mitigating perceptual errors caused by the gap between screen-displayed graphics and the physical world.
- Open Access
- 13 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-3
Data Pictorial: Deconstructing Raster Images for Data-Aware Animated Vector Posters
To support data integration into pictorials, we propose Data Pictorial, a pipeline that deconstructs a raster image into SVG objects whose attributes are contextualized in data. This process is achieved by cropping objects of interest using zero-shot detection, converting them into quantized bitmaps, and tracing the results as SVG paths. The technique then provides suggestions for binding the SVG objects and properties with data fields, affording the flexibility to automatically modify and animate the SVG based on the mapping. The resultant data-aware vector hypermedia can be potential candidates for real-time data inspection and personalization, all while maintaining the aesthetic of the original pictorial.
- 13 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-3
Open AccessData Pictorial: Deconstructing Raster Images for Data-Aware Animated Vector Posters
To support data integration into pictorials, we propose Data Pictorial, a pipeline that deconstructs a raster image into SVG objects whose attributes are contextualized in data. This process is achieved by cropping objects of interest using zero-shot detection, converting them into quantized bitmaps, and tracing the results as SVG paths. The technique then provides suggestions for binding the SVG objects and properties with data fields, affording the flexibility to automatically modify and animate the SVG based on the mapping. The resultant data-aware vector hypermedia can be potential candidates for real-time data inspection and personalization, all while maintaining the aesthetic of the original pictorial.
- Open Access
- 11 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-13
SonifyAR: Context-Aware Sound Generation in Augmented Reality
Sound plays a crucial role in enhancing user experience and immersiveness in Augmented Reality (AR). However, current platforms lack support for AR sound authoring due to limited interaction types, challenges in collecting and specifying context information, and difficulty in acquiring matching sound assets. We present SonifyAR, an LLM-based AR sound authoring system that generates context-aware sound effects for AR experiences. SonifyAR expands the current design space of AR sound and implements a Programming by Demonstration (PbD) pipeline to automatically collect contextual information of AR events, including virtual-content-semantics and real-world context. This context information is then processed by a large language model to acquire sound effects with Recommendation, Retrieval, Generation, and Transfer methods. To evaluate the usability and performance of our system, we conducted a user study with eight participants and created five example applications, including an AR-based science experiment, and an assistive application for low-vision AR users.
- 11 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-13
Open AccessSonifyAR: Context-Aware Sound Generation in Augmented Reality
Sound plays a crucial role in enhancing user experience and immersiveness in Augmented Reality (AR). However, current platforms lack support for AR sound authoring due to limited interaction types, challenges in collecting and specifying context information, and difficulty in acquiring matching sound assets. We present SonifyAR, an LLM-based AR sound authoring system that generates context-aware sound effects for AR experiences. SonifyAR expands the current design space of AR sound and implements a Programming by Demonstration (PbD) pipeline to automatically collect contextual information of AR events, including virtual-content-semantics and real-world context. This context information is then processed by a large language model to acquire sound effects with Recommendation, Retrieval, Generation, and Transfer methods. To evaluate the usability and performance of our system, we conducted a user study with eight participants and created five example applications, including an AR-based science experiment, and an assistive application for low-vision AR users.
- Open Access
- 11 May 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-10
MoiréWidgets: High-Precision, Passive Tangible Interfaces via Moiré Effect
- 11 May 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-10
Open AccessMoiréWidgets: High-Precision, Passive Tangible Interfaces via Moiré Effect