Author name cluster

Ping Li

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.

111 papers

2 author rows

JBHI Journal 2026 Journal Article

Bond-Aware Molecular Graph Learning With Multi-Graph Interleaved Message Passing

Honghao Wang
Hongrui Zhang
Acong Zhang
Junlei Tang
Ping Li

Graph neural networks (GNNs) have demonstrated remarkable capabilities in molecular property prediction. Existing approaches adopt GNNs by modeling molecules as homogeneous graphs. However, the bonds between atoms can be heterogeneous, whose characterization and role in molecular graph representation learning remain unexplored. To address the heterogeneity issue inherent in molecular graphs, in this work, we build the bond-centric graphs and propose a novel multi-graph learning model, which captures the bond heterogeneity via augmented bond graph view and bond coding for atom features. Different from conventional multi-view learning that focus on late-stage view fusion, our method integrates cross-graph information during the node representation learning phase. Towards this end, we introduce the interleaved message passing graph neural network ( IMPGNN ), allowing the messages passing across three views of the molecular graph. Moreover, we introduce a novel structure-aware pooling mechanims for graph representation, which yields up to 45. 7% gains over simple sum pooling. Comparative experiments on two standard molecular property prediction tasks reveal that our method surpasses all competing approaches (including multimodal models) on 75% of the evaluated benchmark datasets.

Details DOI

AAAI Conference 2026 Conference Paper

Discovering Decoupled Functional Modules in Large Language Models

Yanke Yu
Jin Li
Ying Sun
Ping Li
Zhefeng Wang
Yi Zheng

Understanding the internal functional organization of Large Language Models (LLMs) is crucial for improving their trustworthiness and performance. However, how LLMs organize different functions into modules remains highly unexplored. To bridge this gap, we formulate a function module discovery problem and propose an Unsupervised LLM Cross-layer MOdule Discovery (ULCMOD) framework that simultaneously disentangles the large set of neurons in the entire LLM into modules while discovering the topics of input samples related to these modules. Our framework introduces a novel objective function and an efficient Iterative Decoupling (IterD) algorithm. Extensive experiments show that our method discovers high-quality, disentangled modules that capture more meaningful semantic information and achieve superior performance in various downstream tasks. Moreover, our qualitative analysis reveals that the discovered modules show function comprehensiveness, function hierarchy, and clear function spatial arrangement within LLMs. Our work provides a novel tool for interpreting LLMs' function modules, filling a critical gap in LLMs' interpretability research.

PDF Details DOI

EAAI Journal 2026 Journal Article

Elastic net twin support vector machine with Universum data and its safe screening rule

Hongmei Wang
Ping Li
Kun Jiang
Yitian Xu

Twin support vector machine with Universum data (UTSVM) is a powerful classification technique. It not only inherits the sophisticated structure of twin support vector machine but also incorporates prior information from Universum data. However, the adoption of l 1 penalty for slack variables in UTSVM suffers a geometric irrationality, impairing its ability to precisely represent the location of violated samples and thus partially degrading model performance. Therefore, we propose a novel elastic net twin support vector machine with Universum data (ENUTSVM) in this paper, which refines the geometric formulation by imposing elastic net penalty for slack variables. Furthermore, we theoretically derive a violation tolerance upper bound (VTUB) that quantitatively characterizes the relationship between the distances of violated samples and their corresponding slack variable differences. Additionally, to enhance the computational efficiency of ENUTSVM, we develop a safe screening rule (SSR-ENUTSVM) by combining variational inequalities and optimization conditions. We compare the proposed method with seven other competitive algorithms on four synthetic datasets and ten benchmark datasets. The experimental results and statistical tests confirm the superiority of our methods. Finally, we apply our method to an epileptic electroencephalogram (EEG) signal classification problem, which verifies its effectiveness in practical applications.

Details DOI

EAAI Journal 2026 Journal Article

Slimmable neural architecture design based on cross architecture and token distillation

Guhao Qiu
Zhihua Chen
Lei Dai
Ping Li
Bin Sheng

The manually designed neural networks have the drawbacks of requiring a large amount of training data and high computational costs. In this paper, we propose the masked autoencoder based lightweight network search algorithm which leverages the efficient channel search algorithm and specific distillation strategy to obtain the optimal architecture. During SuperNet training process, we design the cross-token distillation and cross-architecture strategy. Token distillation strategy is used to enforce the similar representation obtained from different masks in one image. Architecture distillation strategy is used to fully utilize the representation from the sampled subnetwork and use the feature maps from one image but the same token. In the subnetwork searching process, we further pretrain the selected network considering the component dependency. Comprehensive experiments verify that our proposed method is efficient and flexible than baseline self-supervised learning algorithm and structured pruning algorithms. For example, our method obtains 4. 4% improvement in TOP-1 metrics compared with the classic Masked Autoencoder algorithm designed for lightweight Transformer architecture with less than 10M parameter.

Details DOI

AAAI Conference 2026 Conference Paper

Toward Real-World High-Precision Image Matting and Segmentation

Haipeng Zhou
Zhaohu Xing
Hongqiu Wang
Jun Ma
Ping Li
Lei Zhu

High-precision scene parsing tasks, including image matting and dichotomous segmentation, aim to accurately predict masks with extremely fine details (such as hair). Most existing methods focus on salient, single foreground objects. While interactive methods allow for target adjustment, their class-agnostic design restricts generalization across different categories. Furthermore, the scarcity of high-quality annotation has led to a reliance on inharmonious synthetic data, resulting in poor generalization to real-world scenarios. To this end, we propose a Foreground Consistent Learning model, dubbed as FCLM, to address the aforementioned issues. Specifically, we first introduce a Depth-Aware Distillation strategy where we transfer the depth-related knowledge for better foreground representation. Considering the data dilemma, we term the processing of synthetic data as domain adaptation problem where we propose a domain-invariant learning strategy to focus on foreground learning. To support interactive prediction, we contribute an Object-Oriented Decoder that can receive both visual and language prompts to predict the referring target. Experimental results show that our method quantitatively and qualitatively outperforms state-of-the-art methods.

PDF Details DOI

EAAI Journal 2025 Journal Article

Adaptive human in the loop system for identifying non-optimal states in natural product manufacturing process

Qilong Xue
Yang Yu
Shixin Cen
Yequan Yan
Jiping Pang
Ping Li
Yehan Hou
Lei Wang

In the extraction of natural products, the identification of non-optimal production states is pivotal for ensuring consistent product quality. Currently, there is a deficiency in online, automated detection methods. This study introduces an online machine vision strategy in a real industrial setting to maintain optimal production state. Specifically, the strategy incorporates an adaptive human in the loop deep learning approach to select high-value samples. This method achieves over 90 % accuracy with fewer training samples, effectively addressing the challenges posed by the low-value density characteristic of industrial data. Additionally, a convolutional neural networks-transformer framework is employed as a classifier for video data to meet the demands of time-series data. To enhance the efficiency of processing multiple video streams, we have implemented a knowledge distillation technique to lighten the model. Finally, this model has been deployed in an actual industrial environment for online monitoring of three extraction devices. The system encapsulates the expertise of engineers to standardize the criteria for assessing production states. This integration of innovative technologies ensures a more reliable and efficient extraction process, meeting the industry's need for consistent product quality.

Details DOI

NeurIPS Conference 2025 Conference Paper

Comparator-Adaptive $\Phi$-Regret: Improved Bounds, Simpler Algorithms, and Applications to Games

Soumita Hait
Ping Li
Haipeng Luo
Mengxiao Zhang

In the classic expert problem, $\Phi$-regret measures the gap between the learner's total loss and that achieved by applying the best action transformation $\phi \in \Phi$. A recent work by Lu et al. , [2025] introduced an adaptive algorithm whose regret against a comparator $\phi$ depends on a certain sparsity-based complexity measure of $\phi$, recovering and interpolating optimal bounds for standard regret notions such as external, internal, and swap regret. In this work, we propose a general idea to achieve an even better comparator-adaptive $\Phi$-regret bound via much simpler algorithms compared to Lu et al. , [2025]. Specifically, we discover a prior distribution over all possible binary transformations and show that it suffices to achieve prior-dependent regret against these transformations. Then, we propose two concrete and efficient algorithms to achieve so, where the first one combines multiple copies of the kernelized Hedge algorithm of Farina et al. , [2022], and the second one combines multiple copies of a variant of the BM-reduction [Blum and Mansour, 2007]. To further showcase the power of our methods and the advantages over Lu et al. , [2025] besides the simplicity and better regret bounds, we also show that our second approach can be extended to the game setting to achieve accelerated and adaptive convergence rate to $\Phi$-equilibria for a class of general-sum games. When specified to the special case of correlated equilibria, our bound improves over the existing ones from Anagnostides et al. , [2022a, b].