Author name cluster

Jianbo Yu

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.

6 papers

1 author row

AAAI Conference 2026 Conference Paper

EfficientFSL: Enhancing Few-Shot Classification via Query-Only Tuning In Vision Transformers

Wenwen Liao
Hang Ruan
Jianbo Yu
Bing Song
Yuansong Wang
Xiaofeng Yang

Large models such as Vision Transformers (ViTs) have demonstrated remarkable superiority over smaller architectures like ResNet in few-shot classification, owing to their powerful representational capacity. However, fine-tuning such large models demands extensive GPU memory and prolonged training time, making them impractical for many real-world low-resource scenarios. To bridge this gap, we propose EfficientFSL, a query-only fine-tuning framework tailored specifically for few-shot classification with ViT, which achieves competitive performance while significantly reducing computational overhead. EfficientFSL fully leverages the knowledge embedded in the pre-trained model and its strong comprehension ability, achieving high classification accuracy with an extremely small number of tunable parameters. Specifically, we introduce a lightweight trainable Forward Block to synthesize task-specific queries that extract informative features from the intermediate representations of the pre-trained model in a query-only manner. We further propose a Combine Block to fuse multi-layer outputs, enhancing the depth and robustness of feature representations. Finally, a Support-Query Attention Block mitigates distribution shift by adjusting prototypes to align with the query set distribution. With minimal trainable parameters, EfficientFSL achieves state-of-the-art performance on four in-domain few-shot datasets and six cross-domain datasets, demonstrating its effectiveness in real-world applications.

PDF Details DOI

AAAI Conference 2026 Conference Paper

MicLog: Towards Accurate and Efficient LLM-based Log Parsing via Progressive Meta In-Context Learning

Jianbo Yu
Yixuan Li
Hai Xu
Kang Xu
Junjielong Xu
Zhijing Li
Pinjia He
Wanyuan Wang

Log parsing converts semi-structured logs into structured templates, forming a critical foundation for downstream analysis. Traditional syntax and semantic-based parsers often struggle with semantic variations in evolving logs and data scarcity stemming from their limited domain coverage. Recent large language model (LLM)-based parsers leverage in-context learning (ICL) to extract semantics from examples, demonstrating superior accuracy. However, LLM-based parsers face two main challenges: 1) underutilization of ICL capabilities, particularly in dynamic example selection and cross-domain generalization, leading to inconsistent performance; 2) time-consuming and costly LLM querying. To address these challenges, we present MicLog, the first progressive meta in-context learning (ProgMeta-ICL) log parsing framework that combines meta-learning with ICL on small open-source LLMs (i.e., Qwen-2.5-3B). Specifically, MicLog: i) enhances LLMs' ICL capability through a zero-shot to k-shot ProgMeta-ICL paradigm, employing weighted DBSCAN candidate sampling and enhanced BM25 demonstration selection; ii) accelerates parsing via a multi-level pre-query cache that dynamically matches and refines recently parsed templates. Evaluated on Loghub-2.0, MicLog achieves 10.3% higher parsing accuracy than the state-of-the-art parser while reducing parsing time by 42.4%.

PDF Details DOI

EAAI Journal 2025 Journal Article

Semi-supervised generative adversarial network for plant leaf disease detection

Lixiang Zhao
Jun Hao
Demin Li
Jianbo Yu

Fully supervised plant leaf disease segmentation methods based on convolutional neural networks (CNN) require large quantities of labeled images for training, which are time-consuming and labor-intensive to obtain in practical scenarios. Moreover, most diseases of plant leaf exhibit indistinct edge information and background noise interference, which hinders precise disease localization. To overcome these issues, a semi-supervised generative adversarial network (SSGAN) is proposed for plant leaf disease inspection. Firstly, to learn comprehensive semantic features from limited data, a semi-supervised method based on generative adversarial networks is developed to enhance the interaction of features among labeled and unlabeled disease images. Secondly, a boundary feature attention module (BFAM) is proposed to enhance edge detail representation, which makes the model pay more attention to the boundary feature of plant leaf diseases. Finally, a background noise suppression module (BNSM) is proposed to bolster the differentiation between the normal and disease areas so as to alleviate the adverse impact of background noise. The effectiveness of SSGAN is verified on two plant leaf disease datasets. The mean intersection over union (mIoU) on the apple and tomato leaf datasets improves by 2. 07 % and 2. 28 % respectively, compared with the suboptimal method. The testing results of experiments show that SSGAN achieves great performance on plant leaf disease segmentation under limited labeled data.

Details DOI

EAAI Journal 2024 Journal Article

Residual squeeze-and-excitation convolutional auto-encoder for fault detection and diagnosis in complex industrial processes

Jianbo Yu
Shijin Li
Xing Liu
Hao Li
Mingyan Ma
Peilun Liu
Lichun You

Recently, deep learning has attracted increasing attention in process monitoring. However, it is still a big challenge to extract effective features of process data for fault detection and diagnosis (FDD) by using deep learning-based methods. In this study, a novel deep learning-based model, residual squeeze-and-excitation convolutional auto-encoder (RSECAE) is proposed for FDD in complex industrial processes. A hybrid structure that integrates convolution calculation and auto-encoder is proposed to construct multiple convolution layers and deconvolution layers in the encoder and decoder respectively. Thus, RSECAE can learn effective features from the process signals in an unsupervised manner. At the same time, multiple residual squeeze-and-excitation networks (RSENets) are embedded in the deep network, which effectively preserves the critical feature representations of the convolution layer and transmits them to the decoder in the form of residuals. Finally, maximum mean discrepancy (MMD) term is utilized in RSECAE to reduce the distribution difference between the process signals and the learned features. This enables RSECAE to extract effective features with prior distribution information. RSECAE provides a new solution for process FDD for complicated industrial processes. Two industrial cases are adopted to validate the effectiveness of the RSECAE in process FDD. The experimental results demonstrate the effectiveness and superiority of RSECAE in process fault detection and fault diagnosis.

Details DOI

EAAI Journal 2021 Journal Article

Wafer map defect recognition based on deep transfer learning-based densely connected convolutional network and deep forest

Jianbo Yu
Zongli Shen
Shijin Wang

Due to the complexity and dynamics of the semiconductor manufacturing processes, wafer maps will present various defect patterns caused by various process faults. Identification of those defect patterns on wafer maps can help operators in finding out root-causes of abnormal processes, and then ensures that the manufacturing process is restored to the normal state as soon as possible. This paper proposes a wafer map defect recognition (WMDR) model based on integration of deep transfer learning and deep forest. Firstly, we transfer the network weight parameters of ImageNet to the convolutional neural network (CNN) (i. e. , densely connected convolutional network (DenseNet)) and redesign the classification layer. This reduces the training time and then improves feature learning performance of DenseNet. Moreover, the transfer learning-based feature learning is able to solve class imbalance of wafer defect patterns. Finally, deep forest is utilized to identify the wafer defect pattern based on the abstract features from the wafer maps extracted by DenseNet. The experimental results on an industrial case show that the method can effectively improve WMDR performance and outperforms those well-known CNNs and other typical classifiers.

Details DOI

EAAI Journal 2009 Journal Article

Identifying source(s) of out-of-control signals in multivariate manufacturing processes using selective neural network ensemble

Jianbo Yu
Lifeng Xi
Xiaojun Zhou

In multivariate statistical process control (MSPC), most multivariate quality control charts are shown to be effective in detecting out-of-control signals based upon an overall statistic. But these charts do not relieve the need for pinpointing source(s) of the out-of-control signals. Neural networks (NNs) have excellent noise tolerance and high pattern identification capability in real time, which have been applied successfully in MSPC. This study proposed a selective NN ensemble approach DPSOEN, where several selected NNs are jointly used to classify source(s) of out-of-control signals in multivariate processes. The immediate location of the abnormal source(s) can greatly narrow down the set of possible assignable causes, facilitating rapid analysis and corrective action by quality operators. The performance of DPSOEN is analyzed in multivariate processes. It shows improved generalization performance that outperforms those of single NNs and Ensemble All approach. The investigation proposed a heuristic approach for applying the DPSOEN-based model as an effective and useful tool to identify abnormal source(s) in bivariate statistical process control (SPC) with potential application for MSPC in general.

Details DOI