Author name cluster

Jianhui Wang

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

5 papers

2 author rows

AAAI Conference 2026 Conference Paper

Learning Protein–Ligand Binding in Hyperbolic Space

Jianhui Wang
Wenyu Zhu
Bowen Gao
Xin Hong
Ya-Qin Zhang
Wei-Ying Ma
Yanyan Lan

Protein-ligand binding prediction is central to virtual screening and affinity ranking, two fundamental tasks in drug discovery. While recent retrieval-based methods embed ligands and protein pockets into Euclidean space for similarity-based search, the geometry of Euclidean embeddings often fails to capture the hierarchical structure and fine-grained affinity variations intrinsic to molecular interactions. In this work, we propose HypSeek, a hyperbolic representation learning framework that embeds ligands, protein pockets, and sequences into Lorentz-model hyperbolic space. By leveraging the exponential geometry and negative curvature of hyperbolic space, HypSeek enables expressive, affinity-sensitive embeddings that can effectively model both global activity and subtle functional differences–particularly in challenging cases such as activity cliffs, where structurally similar ligands exhibit large affinity gaps. Our model unifies virtual screening and affinity ranking in a single framework, introducing a protein-guided three-tower architecture to enhance representational structure. HypSeek improves early enrichment in virtual screening on DUD-E from 42.63 to 51.44 (+20.7%) and affinity ranking correlation on JACS from 0.5774 to 0.7239 (+25.4%), demonstrating the benefits of hyperbolic geometry across both tasks and highlighting its potential as a powerful inductive bias for protein-ligand modeling.

PDF Details DOI

EAAI Journal 2025 Journal Article

A unified multi-criteria equity metric for data-driven decision-making: Application to tract-level electric vehicle charging station planning

Zhi Li
Janille Smith-Colin
Jianhui Wang

This paper introduces a unified, data-driven multi-criteria equity metric designed to address inequities in Electric Vehicle Charging Station (EVCS) accessibility and Electric Vehicle (EV) adoption. Existing studies predominantly focus on macro-level factors and rely on single-criterion or context-specific approaches, limiting their applicability to dynamic, multi-dimensional challenges. Additionally, current equity evaluations often involve inherent subjectivity, complicating decision-making processes. Focusing on the Dallas-Fort Worth area, this study advances the field by transitioning from city-level to census tract-level analysis. The proposed framework utilizes dimension reduction techniques to construct a unified equity metric and high-resolution clustering analysis to identify EV-related socio-demographic typologies. These typologies—EV developed, EV developing, and EV underdeveloped/disadvantaged—are validated against disadvantaged communities (DACs) defined by the Climate and Economic Justice Screening Tool (CEJST), mapped EV/EVCS distributions, and their corresponding socio-demographic radar plots. The data-driven equity metric has advantages in unity, objectivity, adaptability, and robustness compared to existing metrics. By identifying micro-level inequities, the framework provides tailored policy recommendations for each typology, supporting equitable EV infrastructure planning.

Details DOI

NeurIPS Conference 2025 Conference Paper

AANet: Virtual Screening under Structural Uncertainty via Alignment and Aggregation

Wenyu Zhu
Jianhui Wang
Bowen Gao
Yinjun Jia
Haichuan Tan
Ya-Qin Zhang
Wei-Ying Ma
Yanyan Lan

Virtual screening (VS) is a critical component of modern drug discovery, yet most existing methods—whether physics-based or deep learning-based—are developed around holo protein structures with known ligand-bound pockets. Consequently, their performance degrades significantly on apo or predicted structures such as those from AlphaFold2, which are more representative of real-world early-stage drug discovery, where pocket information is often missing. In this paper, we introduce an alignment-and-aggregation framework to enable accurate virtual screening under structural uncertainty. Our method comprises two core components: (1) a tri-modal contrastive learning module that aligns representations of the ligand, the holo pocket, and cavities detected from structures, thereby enhancing robustness to pocket localization error; and (2) a cross-attention based adapter for dynamically aggregating candidate binding sites, enabling the model to learn from activity data even without precise pocket annotations. We evaluated our method on a newly curated benchmark of apo structures, where it significantly outperforms state-of-the-art methods in blind apo setting, improving the early enrichment factor (EF1\%) from 11. 75 to 37. 19. Notably, it also maintains strong performance on holo structures. These results demonstrate the promise of our approach in advancing first-in-class drug discovery, particularly in scenarios lacking experimentally resolved protein-ligand complexes. Our implementation is publicly available at https: //github. com/Wiley-Z/AANet.

PDF Details

EAAI Journal 2025 Journal Article

Precise step counting algorithm for pedestrians using ultra-low-cost foot-mounted accelerometer

Jingxue Bi
Jianhui Wang
Baoguo Yu
Guobiao Yao
Yunjia Wang
Hongji Cao
Lu Huang
Huaqiao Xing

Zero-velocity update (ZUPT) is one of the most widely used step counting methods in pedestrian inertial navigation. Existing accelerometer-based methods encounter issues with misjudgments of zero velocity and step counting errors in various activities. To tackle these challenges, we present a precise step counting algorithm leveraging ultra-low-cost foot-mounted accelerometer for pedestrians with extremely low complexity. This algorithm mainly contains two key points: adaptive acceleration threshold selection and state vector update, implemented by twice zero-velocity detections (ZVD, generating state vectors to depict pedestrian's status) and length comparisons. Accelerations undergo gravity correction, magnitude calculation, data smoothing, and parameters initialization. Subsequently, the zero-acceleration threshold is adaptively selected through ZVD, and the length of state intervals (LSI) is compared with the first length threshold. Then, the state vector is updated by ZVD, and the LSI is compared with the second length threshold. Finally, the number of state intervals is interpreted as step counts. Step counting experiments were conducted utilizing three diverse datasets, which comprised a self-constructed ultra-low-cost accelerometer-based dataset and two public datasets. These datasets covered various activities, such as normal walking, fast walking, running, multiple turns in a corridor, stationary stepping, and upstairs/downstairs. The proposed algorithm could achieve an accuracy of 100 % under the above situations. Compared to long short term memory (LSTM) and other algorithms, it exhibited an accuracy improvement of at least 13. 05 %, with a processing time only 0. 07 % of that required by LSTM. The proposed algorithm enables accurate step counting across a range of activities performed by different pedestrians with high precision, low complexity, and robust applicability.

Details DOI

ICRA Conference 2025 Conference Paper

Wcdt: World-Centric Diffusion Transformer for Traffic Scene Generation

Chen Yang
Yangfan He
Aaron Xuxiang Tian
Dong Chen 0016
Jianhui Wang
Tianyu Shi
Arsalan Heydarian
Pei Liu

In this paper, we introduce a novel approach for autonomous driving trajectory generation by harnessing the complementary strengths of diffusion probabilistic models (a. k. a. , diffusion models) and transformers. Our proposed framework, termed the “World-centric Diffusion Transformer” (WcDT), optimizes the entire trajectory generation process, from feature extraction to model inference. To enhance the scene diversity and stochasticity, the historical trajectory data is first preprocessed into “Agent Move Statement” and encoded into latent space using Denoising Diffusion Probabilistic Models (DDPM) enhanced with Diffusion with Transformer (DiT) blocks. Then, the latent features, historical trajectories, HD map features, and historical traffic signal information are fused with various transformer-based encoders that is used to enhance the interaction of agents with other elements in the traffic scene. The encoded traffic scenes are then decoded by a trajectory decoder to generate multimodal future trajectories. Comprehensive experimental results show that the proposed approach exhibits superior performance in generating both realistic and diverse trajectories, showing its potential for integration into automatic driving simulation systems. Our code is available at https://github.com/yangchen1997/WcDT.

Details