Le Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Le Liu has a total of 41 co-authors. Here are the top co-authors by number of shared publications:
- Peng Wang
Northwestern Polytechnical University, Xi'an, China
4 shared publications - Di Xu
Huawei Technologies Co., Ltd., Xi'an, China
3 shared publications - Shizhou Zhang
Northwestern Polytechnical University, Xi'an, China
3 shared publications - Fugee Tsung
The Hong Kong University of Science and Technology, Hong Kong
2 shared publications - Kaixing Zhao
Northwestern Polytechnical University
2 shared publications - Liangwei Wang
The Hong Kong University of Science and Technology (Guangzhou), Guangzhou, China
2 shared publications
Latest Publications
- Open Access
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 9382-9391
PBU-ReCab: Partial Backbone Unfreezing and Retrospective Calibration for Domain-Incremental Object Detection
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 9382-9391
Open AccessPBU-ReCab: Partial Backbone Unfreezing and Retrospective Calibration for Domain-Incremental Object Detection
- Open Access
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 17066-17073
MoRe-UAV: A Large-Scale Benchmark for Motion-Aware Visual Grounding in UAV Videos
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 17066-17073
Open AccessMoRe-UAV: A Large-Scale Benchmark for Motion-Aware Visual Grounding in UAV Videos
- Open Access
- 18 December 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-5
VisualNetO&M: A Digital Twin-Based Collaborative Visualization System for Power System Communication Network Operation and Maintenance
- 18 December 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-5
Open AccessVisualNetO&M: A Digital Twin-Based Collaborative Visualization System for Power System Communication Network Operation and Maintenance
- Open Access
- 18 December 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-8
Touch, Sound, and Space: Exploring Immersive Music Interaction through AI-Generated Environments
- 18 December 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-8
Open AccessTouch, Sound, and Space: Exploring Immersive Music Interaction through AI-Generated Environments
- Open Access
- 18 December 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-5
ChatHSI: Reliable LLM-Powered Human-Swarm Interaction Framework
- 18 December 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-5
Open AccessChatHSI: Reliable LLM-Powered Human-Swarm Interaction Framework
- Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 8882-8890
SUVIS: A Depth- and Motion-Encoded Stereoscopic System for Communicating Forecast Uncertainty
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 8882-8890
Open AccessSUVIS: A Depth- and Motion-Encoded Stereoscopic System for Communicating Forecast Uncertainty
- 23 May 2025
- Published by Wiley in Computer Graphics Forum
- Vol. 44 (3)
VizTA: Enhancing Comprehension of Distributional Visualization with Visual‐Lexical Fused Conversational Interface
Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM‐based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine‐grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic‐aware conversational agent capable of explaining contextual information within visualizations and employs a visual‐lexical fusion design to facilitate chart‐centered conversation. A between‐subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.
- 23 May 2025
- Published by Wiley in Computer Graphics Forum
- Vol. 44 (3)
VizTA: Enhancing Comprehension of Distributional Visualization with Visual‐Lexical Fused Conversational Interface
Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM‐based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine‐grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic‐aware conversational agent capable of explaining contextual information within visualizations and employs a visual‐lexical fusion design to facilitate chart‐centered conversation. A between‐subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.
- Open Access
VizTA: Enhancing Comprehension of Distributional Visualization with Visual-Lexical Fused Conversational Interface
Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.
Open AccessVizTA: Enhancing Comprehension of Distributional Visualization with Visual-Lexical Fused Conversational Interface
Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.
- Open Access
- 8 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
A Representative Sampling Strategy for Implicit Uncertainty Visualization of Multidimensional Ensembles
Ensemble datasets are commonly used in predictive modeling and scientific simulations to represent uncertainty and variability in complex systems. However, effectively visualizing ensemble data while preserving its inherent uncertainty remains challenging due to multidimensionality, visual clutter, and multimodal distributions. Traditional visualization methods, including statistical summaries and direct displays of ensembles, often struggle to balance interpretability with fidelity, leading to potential biases or the loss of important structural information. This paper presents a novel representative sampling strategy designed for multidimensional spaces that enhances ensemble visualization by preserving uncertainty characteristics while reducing complexity. Our approach first applies Centroidal Voronoi Tessellation to achieve a uniform data distribution while maintaining topological consistency. Second, we implement Weighted Sample Elimination to extract a subset of representative samples that adhere to the Poisson disk property. This subset is the foundation for visualizations, ensuring distribution preservation and reduced redundancy. By refining ensemble representations, our method improves visual clarity while maintaining the statistical integrity of the original dataset. We validate our approach using a variety of multidimensional ensembles, including hurricane predictions, medical tractography, and synthetic ensembles. The results show that our sampled subsets effectively preserve essential distributional features while significantly reducing visual clutter. Our work lays a strong foundation for uncertainty-aware visualization techniques and opens new opportunities for scalable, high-fidelity ensemble analysis across scientific and engineering fields.
- 8 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
Open AccessA Representative Sampling Strategy for Implicit Uncertainty Visualization of Multidimensional Ensembles
Ensemble datasets are commonly used in predictive modeling and scientific simulations to represent uncertainty and variability in complex systems. However, effectively visualizing ensemble data while preserving its inherent uncertainty remains challenging due to multidimensionality, visual clutter, and multimodal distributions. Traditional visualization methods, including statistical summaries and direct displays of ensembles, often struggle to balance interpretability with fidelity, leading to potential biases or the loss of important structural information. This paper presents a novel representative sampling strategy designed for multidimensional spaces that enhances ensemble visualization by preserving uncertainty characteristics while reducing complexity. Our approach first applies Centroidal Voronoi Tessellation to achieve a uniform data distribution while maintaining topological consistency. Second, we implement Weighted Sample Elimination to extract a subset of representative samples that adhere to the Poisson disk property. This subset is the foundation for visualizations, ensuring distribution preservation and reduced redundancy. By refining ensemble representations, our method improves visual clarity while maintaining the statistical integrity of the original dataset. We validate our approach using a variety of multidimensional ensembles, including hurricane predictions, medical tractography, and synthetic ensembles. The results show that our sampled subsets effectively preserve essential distributional features while significantly reducing visual clutter. Our work lays a strong foundation for uncertainty-aware visualization techniques and opens new opportunities for scalable, high-fidelity ensemble analysis across scientific and engineering fields.
- Open Access
- 11 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-8
Evaluating the Design Effectiveness of Radial Layout Glyph Visualizations for Multivariate Data: A Perception Study
- 11 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-8
Open AccessEvaluating the Design Effectiveness of Radial Layout Glyph Visualizations for Multivariate Data: A Perception Study