Author name cluster

Chuan Lin

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.

11 papers

1 author row

AAAI Conference 2026 Conference Paper

A Reasoning Paradigm for Named Entity Recognition

Hui Huang
Yanping Chen
Ruizhang Huang
Chuan Lin
Yongbin Qin

Generative LLMs typically improve Named Entity Recognition (NER) performance through instruction tuning. They excel at generating entities by semantic pattern matching but lack an explicit, verifiable reasoning mechanism. This "cognitive shortcutting" leads to suboptimal performance and weak generalization, especially in zero-shot and low-resource scenarios where reasoning from limited contextual cues is crucial. To address this issue, a reasoning framework is proposed for NER, which shifts the extraction paradigm from implicit pattern matching to explicit reasoning. This framework consists of three stages: Chain of Thought (CoT) generation, CoT tuning, and reasoning enhancement. First, a dataset annotated with NER-oriented CoTs is generated, which contain task-relevant reasoning chains. Then, they are used to tune the NER model to generate coherent rationales before deriving the final answer. Finally, a reasoning enhancement stage is implemented to optimize the reasoning process using a comprehensive reward signal. This stage ensures explicit and verifiable extractions. Experiments show that ReasoningNER demonstrates impressive cognitive ability in the NER task, achieving competitive performance. In zero-shot settings, it achieves SoTA performance, outperforming GPT-4 by 12.3 percentage points on the F1 score. Analytical results demonstrate its great potential to advance research in reasoning-oriented information extraction.

PDF Details DOI

JBHI Journal 2026 Journal Article

CoMIL: A Contrastive CNN-Transformer Framework with Multi-Instance Learning for Whole-Slide Pathology Image Classification

Bowen Liu
Hongbo Zhu
Xiaotong Wei
Chuan Lin
Wei Wang

Whole slide image (WSI) classification faces challenges due to gigapixel scale and weak supervision, often struggling to balance global context with local details. We propose CoMIL, a dual-branch framework based on symmetric mutual learning. Firstly, to resolve the dilemma where single-stream networks struggle to simultaneously capture global context and fine-grained details, we employ dual parallel pathways: a Transformer branch models long-range instance dependencies, while a CNN branch captures localized tissue morphology. Secondly, to address spatial information loss, we design a Hyper Positional Generator (HyperPG). This module integrates multi-scale adaptive mechanisms with deformable convolutions, enhancing spatial awareness with linear complexity. Finally, to improve model robustness against weak label noise, bidirectional learning between branches is achieved through KL divergence minimization. Extensive experiments show that our proposed method achieves an area under the curve of 98. 6% and an accuracy of 95. 3% on the Camelyon16 dataset, and an area under the curve of 98. 8% and an accuracy of 93. 3% on the TCGA_Kidney dataset, surpassing the performance of known advanced WSI classification methods.

Details DOI

EAAI Journal 2025 Journal Article

ESIGCF: Extremely simplified but intent-enhanced graph collaborative filtering for recommendation

Zhi Yang
Ruizhang Huang
Yanping Chen
Chuan Lin
Yongbin Qin

Graph convolutional networks (GCNs) and graph contrastive learning (GCL) have substantially advanced recommender systems by modeling high-order user–item interactions and leveraging self-supervised signals. However, many existing methods overemphasize user–user or item–item similarities and rely on complex intent modeling, leading to increased complexity and limited exposure to diverse items. To address these challenges, we propose ESIGCF (Extremely Simplified but Intent-enhanced Graph Collaborative Filtering) — a lightweight yet effective recommendation framework. ESIGCF explicitly defines user intent as the inner product between user and item embedding vectors and comprises two primary modules: (i) an intent-enhanced GCN that uses hybrid normalization (combining mean- and symmetric-normalization) to capture fine-grained user–item preferences without additional intent parameters, and (ii) an intent-aware GCL that aligns user–item pairs and positive and generated negative items. Negative samples are generated via a non-linear activation of item embedding interactions, promoting exposure to varied candidates without data augmentation. Experiments on three public datasets (Alibaba-iFashion, Yelp2018, Amazon-Book) show that ESIGCF consistently outperforms state-of-the-art baselines. For instance, on Alibaba-iFashion, ESIGCF achieves Recall@20 of 0. 1273 versus 0. 1059 for the best intent-enhanced baseline (a 20. 2% relative improvement). Comprehensive experiments confirm that ESIGCF effectively captures latent user intent, mitigates popularity bias, and enhances recommendation performance with reduced complexity. Our code is available at https: //github. com/Yangzhi22/ESIGCF.

Details DOI

EAAI Journal 2025 Journal Article

Real-time inner wall surface defect detection based on multi-morphological feature fusion network

Zhenshen Qu
Xinxu Cai
Tianyi Zhang
Jiazheng Xu
Xintong Jiang
Chuan Lin

In industrial manufacturing, defects on the inner wall surface are crucial for quality and safety assessment. However, existing detection methods are limited by low resolution and glare interference. This study presents a Multi-morphological Feature Fusion Network for Object Detection (MFFN-OD) for 360°detection of inner wall image defects. First, it cleverly integrates panoramic imaging with conventional features through a dual-branch backbone and annular features, ensuring rotation invariance and holistic feature preservation. Second, we develop an Adaptive Multi-morphological Feature Alignment Module (AMFAM) that combats centrally polarized defects by automatically adjusting feature alignment, reducing noise, and increasing accuracy, as well as a feature interaction module with a focus on strengthening multiscale feature fusion. Third, we introduce an Asymptotic Feature Pyramid Network with Auxiliary Features (AFPN-AF) to further refine fusion, close semantic gaps, and improve performance. Experimental results show that MFFN-OD achieves 96. 1% mean Average Precision (mAP) and 94. 3% Average Precision (AP) for demanding faults with fast detection of 17 milliseconds per frame, meeting industrial requirements for accuracy and real-time performance.

Details DOI

AIJ Journal 2024 Journal Article

A multi-graph representation for event extraction

Hui Huang
Yanping Chen
Chuan Lin
Ruizhang Huang
Qinghua Zheng
Yongbin Qin

Event extraction has a trend in identifying event triggers and arguments in a unified framework, which has the advantage of avoiding the cascading failure in pipeline methods. The main problem is that joint models usually assume a one-to-one relationship between event triggers and arguments. It leads to the argument multiplexing problem, in which an argument mention can serve different roles in an event or shared by different events. To address this problem, we propose a multigraph-based event extraction framework. It allows parallel edges between any nodes, which is effective to represent semantic structures of an event. The framework enables the neural network to map a sentence(s) into a structurized semantic representation, which encodes multi-overlapped events. After evaluated on four public datasets, our method achieves the state-of-the-art performance, outperforming all compared models. Analytical experiments show that the multigraph representation is effective to address the argument multiplexing problem and helpful to advance the discriminability of the neural network for event extraction.

Details DOI

JBHI Journal 2024 Journal Article

EEG-Based Mental Workload Classification Method Based on Hybrid Deep Learning Model Under IoT

Shiliang Shao
Guangjie Han
Ting Wang
Chuan Lin
Chunhe Song
Chen Yao

Automatically detecting human mental workload to prevent mental diseases is highly important. With the development of information technology, remote detection of mental workload is expected. The development of artificial intelligence and Internet of Things technology will also enable the identification of mental workload remotely based on human physiological signals. In this article, a method based on the spatial and time–frequency domains of electroencephalography (EEG) signals is proposed to improve the classification accuracy of mental workload. Moreover, a hybrid deep learning model is presented. First, the spatial domain features of different brain regions are proposed. Simultaneously, EEG time–frequency domain information is obtained based on wavelet transform. The spatial and time–frequency domain features are input into two types of deep learning models for mental workload classification. To validate the performance of the proposed method, the Simultaneous Task EEG Workload public database is used. Compared with the existing methods, the proposed approach shows higher classification accuracy. It provides a novel means of assessing mental workload.

Details DOI

AILAW Journal 2024 Journal Article

Self-training improves few-shot learning in legal artificial intelligence tasks

Yulin Zhou
Yongbin Qin
Ruizhang Huang
Yanping Chen
Chuan Lin
Yuan Zhou

Abstract As the labeling costs in legal artificial intelligence tasks are expensive. Therefore, it becomes a challenge to utilize low cost to train a robust model. In this paper, we propose a LAIAugment approach, which aims to enhance the few-shot learning capability in legal artificial intelligence tasks. Specifically, we first use the self-training approach to label the amount of unlabelled data to enhance the feature learning capability of the model. Moreover, we also search for datasets that are similar to the training set by improving the text similarity function. We conducted experimental analyses for three legal artificial intelligence tasks, including evidence extraction, legal element extraction, and case multi-label prediction, which composed of 3500 judgement documents. The experimental results show that the proposed LAIAugment method has an average F1-score of 72. 3% on the three legal AI tasks, which is 1. 93% higher than the baseline model. At the same time, it shows a huge improvement in few-shot learning.

Details DOI

AILAW Journal 2023 Journal Article

A novel MRC framework for evidence extracts in judgment documents

Yulin Zhou
Lijuan Liu
Yanping Chen
Ruizhang Huang
Yongbin Qin
Chuan Lin

Abstract Evidences are important proofs to support judicial trials. Automatically extracting evidences from judgement documents can be used to assess the trial quality and support “Intelligent Court”. Current evidence extraction is primarily depended on sequence labelling models. Despite their success, they can only assign a label to a token, which is difficult to recognize nested evidence entities in judgment documents, where a token may belong to several evidences at the same time. In this paper, we present a novel evidence extraction architecture called ATT-MRC, in which extracting evidence entities is formalized as a question answer problem, where all evidence spans are screened out as possible correct answers. Furthermore, to address the data imbalance problem in the judgement documents, we revised the loss function and combined it with a data enhancement technique. Experimental results demonstrate that our model has better performance than related works in evidence extraction.

Details DOI

EAAI Journal 2023 Journal Article

BLEDNet: Bio-inspired lightweight neural network for edge detection

Zhengqiao Luo
Chuan Lin
Fuzhang Li
Yongcai Pan

Edge detection is fundamental to advanced computer vision tasks. Although deep learning-based methods can generate excellent results, they tend to be computationally expensive. To address this issue, researchers have explored the development of lightweight CNNs. Physiological studies have shown that there are two visual pathways in the biological visual system. The second visual pathway can modulate the first visual pathway, aiding the visual cortex in extracting edges from images more quickly. The inferior temporal cortex can integrate the edge signals extracted by the visual pathways and generate edges. Inspired by this, we designed a bio-inspired lightweight edge detection CNN, which includes an Encoder and a Decoder. The Encoder consists of a bio-inspired first visual pathway network and a bio-inspired second visual pathway network. The Decoder consists of a compact receptive field enhanced network. Specifically, the bio-inspired first visual pathway network can calculate the rate of change of image pixels. Through adaptive antagonism, signals with different rates of change are mutually suppressed. Meanwhile, the bio-inspired second visual pathway network can guide the bio-inspired first visual pathway network to focus more on the salient parts of the image. In addition, the receptive field-enhanced network can perceive image features distributed along specific directions, helping the decoder generate clearer edges. Our proposed method achieves competitive results on the BSDS500 dataset with ODS=0. 805 and on the NYUD-v2 dataset with ODS = 0. 757, while also having lower computational costs than most existing methods. Therefore, our approach is suitable for devices with lower computing capabilities.

Details DOI

EAAI Journal 2023 Journal Article

MPDNet: An underwater image deblurring framework with stepwise feature refinement module

Guangjie Han
Min Wang
Hongbo Zhu
Chuan Lin

In this study, a general network model called multi-progressive image deblurring network is proposed to correct blurring artifacts and local imaging details in underwater images. As a solution to nonuniform image distortion, a deformable convolution module was designed to enrich the encoded information of the image representation. Using a stepwise feature refinement module, multi-progressive image deblurring network can reduce the loss of contextual information to produce a more realistic underwater image for subsequent applications. Constructing a loss function based on multi-scale content can help the model improve image perception quality. We conducted experimental evaluations on large-scale image deblurring benchmark datasets, such as GoPro and HIDE, achieving excellent results with 32. 84 dB and 31. 03 dB peak signal-to-noise ratio, respectively, using the proposed method. Subsequently, a detailed optimization comparison was conducted on the in-house underwater image deblurring dataset. Multi-progressive image deblurring network obtained higher-quality, clearer images. Compared with the current state-of-the-art image deblurring algorithms, the proposed model achieved significant results with a 6. 6% increase in deblur performance in peak signal-to-noise ratio. Finally, we conducted ablation experiments to evaluate the effectiveness of all the modules in the proposed framework.

Details DOI

JBHI Journal 2021 Journal Article

Two-Way MR-Forest Based Growing Path Classification for Malignancy Estimation of Pulmonary Nodules

Hongbo Zhu
Guangjie Han
Chuan Lin
Min Wang
Mohsen Guizani
Jianxia Hou
Wei Xing

This paper proposes a two-way multi-ringed forest (TMR-Forest) to estimating the malignancy of the pulmonary nodules for false positive reduction (FPR). Based on our previous work of deep decision framework, named MR-Forest, we generate a growing path mode on predefined pseudo-timeline of $L$ time slots to build pseudo-spatiotemporal features. It synchronously works with FPR based on MR-Forest to help predict the labels from a dynamic perspective. Concretely, Mask R-CNN is first used to recommend the bounding boxes of ROIs and classify their pathological features. Afterward, hierarchical attribute matching is introduced to obtain the input ROIs’ attribute layouts and select the candidates for their growing path generation. The selected ROIs can replace the fixed-sized ROIs’ fitting results at different time slots for data augmentation. A two-stage counterfactual path elimination is used to screen out the input paths of the cascade forest. Finally, a simple label selection strategy is executed to output the predicted label to point out the input nodule's malignancy. On 1034 scans of the merged dataset, the framework can report more accurate malignancy labels to achieve a better CPM score of 0. 912, which exceeds those of MR-Forest and 3DDCNNs about 2. 8% and 4. 7%, respectively.

Details DOI