Author name cluster

Yucheng 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.

15 papers

2 author rows

AAAI Conference 2026 Conference Paper

A Unified Shape-Aware Foundation Model for Time Series Classification

Zhen Liu
Yucheng Wang
Boyuan Li
Junhao Zheng
Emadeldeen Eldele
Min Wu
Qianli Ma

Foundation models pre-trained on large-scale source datasets are reshaping the traditional training paradigm for time series classification. However, existing time series foundation models primarily focus on forecasting tasks and often overlook classification-specific challenges, such as modeling interpretable shapelets that capture class-discriminative temporal features. To bridge this gap, we propose UniShape, a unified shape-aware foundation model designed for time series classification. UniShape incorporates a shape-aware adapter that adaptively aggregates multiscale discriminative subsequences (shapes) into class tokens, effectively selecting the most relevant subsequence scales to enhance model interpretability. Meanwhile, a prototype-based pretraining module is introduced to jointly learn instance- and shape-level representations, enabling the capture of transferable shape patterns. Pre-trained on a large-scale multi-domain time series dataset comprising 1.89 million samples, UniShape exhibits superior generalization across diverse target domains. Experiments on 128 UCR datasets and 30 additional time series datasets demonstrate that UniShape achieves state-of-the-art classification performance, with interpretability and ablation analyses further validating its effectiveness.

PDF Details DOI

AAAI Conference 2026 Conference Paper

BEE-RAG: Balanced Entropy Engineering for Retrieval-Augmented Generation

Yuhao Wang
Ruiyang Ren
Yucheng Wang
Jing Liu
Xin Zhao
Hua Wu
Haifeng Wang

With the rapid advancement of large language models (LLMs), retrieval-augmented generation (RAG) has emerged as a critical approach to supplement the inherent knowledge limitations of LLMs. However, due to the typically large volume of retrieved information, RAG tends to operate with long context lengths. From the perspective of entropy engineering, we identify unconstrained entropy growth and attention dilution due to long retrieval context as significant factors affecting RAG performance. In this paper, we propose the balanced entropy-engineered RAG (BEE-RAG) framework, which improves the adaptability of RAG systems to varying context lengths through the principle of entropy invariance. By leveraging balanced context entropy to reformulate attention dynamics, BEE-RAG separates attention sensitivity from context length, ensuring a stable entropy level. Building upon this, we introduce a zero-shot inference strategy for multi-importance estimation and a parameter-efficient adaptive fine-tuning mechanism to obtain the optimal balancing factor for different settings. Extensive experiments across multiple RAG tasks demonstrate the effectiveness of BEE-RAG.

PDF Details DOI

AAAI Conference 2026 Conference Paper

MonoDream: Monocular Vision-Language Navigation with Panoramic Dreaming

Shuo Wang
Yongcai Wang
Zhaoxin Fan
Yucheng Wang
Maiyue Chen
Kaihui Wang
Zhizhong Su
Wanting Li

Vision-Language Navigation (VLN) tasks often leverage panoramic RGB and depth inputs to provide rich spatial cues for action planning, but these sensors can be costly or less accessible in real-world deployments. Recent approaches based on Vision-Language Action (VLA) models achieve strong results with monocular input, yet they still lag behind methods using panoramic RGB-D information. We present MonoDream, a lightweight VLA framework that enables monocular agents to learn a Unified Navigation Representation (UNR). This shared feature representation jointly aligns navigation-relevant visual semantics (e.g., global layout, depth, and future cues) and language-grounded action intent, enabling more reliable action prediction. MonoDream further introduces Latent Panoramic Dreaming (LPD) tasks to supervise the UNR, which train the model to predict latent features of panoramic RGB and depth observations at both current and future steps based on only monocular input. Experiments on multiple VLN benchmarks show that MonoDream consistently improves monocular navigation performance and significantly narrows the gap with panoramic-based agents.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

Aux-Think: Exploring Reasoning Strategies for Data-Efficient Vision-Language Navigation

Shuo Wang
Yongcai Wang
Wanting Li
Xudong Cai
Yucheng Wang
Maiyue Chen
Zhizhong Su
Deying Li

Vision-Language Navigation is a critical task for developing embodied agents that can follow natural language instructions to navigate in complex real-world environments. Recent advances by finetuning large pretrained models have significantly improved generalization and instruction grounding compared to traditional approaches. However, the role of reasoning strategies in navigation—an action-centric, long-horizon task—remains underexplored, despite Chain-of-Thought reasoning's demonstrated success in static tasks like question answering and visual reasoning. To address this gap, we conduct the first systematic evaluation of reasoning strategies for VLN, including No-Think (direct action prediction), Pre-Think (reason before action), and Post-Think (reason after action). Surprisingly, our findings reveal the Inference-time Reasoning Collaps issue, where inference-time reasoning degrades navigation accuracy, highlighting the challenges of integrating reasoning into VLN. Based on this insight, we propose Aux-Think, a framework that trains models to internalize structured reasoning patterns through CoT supervision during training, while preserving No-Think inference for efficient action prediction. To support this framework, we release R2R-CoT-320k, a large-scale Chain-of-Thought annotated dataset. Empirically, Aux-Think significantly reduces training effort without compromising performance.

PDF Details

NeurIPS Conference 2025 Conference Paper

C-LoRA: Contextual Low-Rank Adaptation for Uncertainty Estimation in Large Language Models

Amir Hossein Rahmati
Sanket Jantre
Weifeng Zhang
Yucheng Wang
Byung-Jun Yoon
Nathan Urban
Xiaoning Qian

Low-Rank Adaptation (LoRA) offers a cost-effective solution for fine-tuning large language models (LLMs), but it often produces overconfident predictions in data-scarce few-shot settings. To address this issue, several classical statistical learning approaches have been repurposed for scalable uncertainty-aware LoRA fine-tuning. However, these approaches neglect how input characteristics affect the predictive uncertainty estimates. To address this limitation, we propose Contextual Low-Rank Adaptation ( C-LoRA ) as a novel uncertainty-aware and parameter efficient fine-tuning approach, by developing new lightweight LoRA modules contextualized to each input data sample to dynamically adapt uncertainty estimates. Incorporating data-driven contexts into the parameter posteriors, C-LoRA mitigates overfitting, achieves well-calibrated uncertainties, and yields robust predictions. Extensive experiments on LLaMA2-7B models demonstrate that C-LoRA consistently outperforms the state-of-the-art uncertainty-aware LoRA methods in both uncertainty quantification and model generalization. Ablation studies further confirm the critical role of our contextual modules in capturing sample-specific uncertainties. C-LoRA sets a new standard for robust, uncertainty-aware LLM fine-tuning in few-shot regimes. Although our experiments are limited to 7B models, our method is architecture-agnostic and, in principle, applies beyond this scale; studying its scaling to larger models remains an open problem. Our code is available at https: //github. com/ahra99/c_lora.

PDF Details

EAAI Journal 2025 Journal Article

Liquid metal microfluidic cooling system for high-efficiency thermal management via learning-based genetic algorithm

Yucheng Wang
Antong Bi
Kaiyu Chen
Shenxin Yu
Wanping Gao
Wenyi Zhang
Yuwan Wu
Zhiqiang Li

High heat flux density is a critical factor that limits the performance and reliability of miniaturized, high-power microelectronic systems. This study proposes a liquid metal (LM)-based microfluidic cooling system optimized through a data-driven computational framework based on an enhanced Genetic Algorithm (LC-GA), aiming to deliver an efficient thermal management solution for high-density integrated systems. By integrating LM near-junction cooling with microchannel heat dissipation in a silicon substrate, we developed a heterogeneous three-dimensional interconnect cooling architecture capable of optimizing thermal performance through algorithm-guided parameter tuning. To validate the proposed method, four distinct microchannel configurations were designed, fabricated, and experimentally tested. LM was introduced into the channels to conduct both experimental cooling tests and thermal performance simulations on a simulated heat source. The results demonstrate that this LM-based microfluidic cooling system, optimized through computational parameter determination, can effectively dissipate heat from chips with power consumption up to 800 W while maintaining stable thermal performance. Additionally, a response surface methodology combined with enhanced LC-GA was utilized for multi-factor sensitivity analysis and multi-objective optimization, enabling automatic determination of optimal design and operating parameters to balance thermal resistance and pressure drop. The optimized configuration reduced the maximum chip temperature to approximately 357. 54 K, lowered the system pressure requirement, and improved the Performance Evaluation Criterion (PEC) to 2. 327. This work provides a data-driven optimization approach that supports the development of high-performance integrated microsystems through algorithm-assisted thermal design.

Details DOI

IROS Conference 2024 Conference Paper

Conditional Generative Denoiser for Nighttime UAV Tracking

Yucheng Wang
Changhong Fu 0001
Kunhan Lu
Liangliang Yao
Haobo Zuo

State-of-the-art (SOTA) visual object tracking methods have significantly enhanced the autonomy of unmanned aerial vehicles (UAVs). However, in low-light conditions, the presence of irregular real noise from the environments severely degrades the performance of these SOTA methods. Moreover, existing SOTA denoising techniques often fail to meet the real-time processing requirements when deployed as plug-and-play denoisers for UAV tracking. To address this challenge, this work proposes a novel conditional generative denoiser (CG-Denoiser), which breaks free from the limitations of traditional deterministic paradigms and generates the noise conditioning on the input, subsequently removing it. To better align the input dimensions and accelerate inference, a novel nested residual Transformer conditionalizer is developed. Furthermore, an innovative multi-kernel conditional refiner is designed to pertinently refine the denoised output. Extensive experiments show that CGDenoiser promotes the tracking precision of the SOTA tracker by 18. 18% on DarkTrack2021 whereas working 5. 8 times faster than the second well-performed denoiser. Real-world tests with complex challenges also prove the effectiveness and practicality of CGDenoiser. Code, video demo and supplementary proof for CGDenoier are now available at: https://github.com/vision4robotics/CGDenoiser.

Details

ICRA Conference 2024 Conference Paper

Enhancing Tactile Sensing in Robotics: Dual-Modal Force and Shape Perception with EIT-based Sensors and MM-CNN

Haofeng Chen
Xuanxuan Yang
Gang Ma 0008
Yucheng Wang
Xiaojie Wang 0004

Electrical Impedance Tomography (EIT)-based tactile sensors offer durability, scalability, and cost-effective manufacturing. However, simultaneously reconstructing force and shape from boundary measurements remains challenging due to EIT’s inherent location dependencies and image artifacts. This study presents a model-driven multimodal convolutional neural network (MM-CNN) for joint EIT-based force and shape sensing. The hybrid approach combines physics-inspired voltage preprocessing with an attention-based network to overcome EIT’s limitations. The preprocessing network applies a linearized one-step inverse solution with Tikhonov regularization to convert raw boundary voltage into a noise-reduced 2D image. The image reconstruction network uses an attention mechanism to focus on salient features, addressing location dependency issues. Quantitative metrics show that MM-CNN outperforms traditional EIT algorithms like NOSER and TV, reducing location dependency and improving shape discrimination. MM-CNN enables unified force and shape modalities, validated through real-contact experiments, enhancing EIT tactile systems for human-robot interaction by incorporating physical knowledge with deep learning.

Details

AAAI Conference 2024 Conference Paper

Fully-Connected Spatial-Temporal Graph for Multivariate Time-Series Data

Yucheng Wang
Yuecong Xu
Jianfei Yang
Min Wu
Xiaoli Li
Lihua Xie
Zhenghua Chen

Multivariate Time-Series (MTS) data is crucial in various application fields. With its sequential and multi-source (multiple sensors) properties, MTS data inherently exhibits Spatial-Temporal (ST) dependencies, involving temporal correlations between timestamps and spatial correlations between sensors in each timestamp. To effectively leverage this information, Graph Neural Network-based methods (GNNs) have been widely adopted. However, existing approaches separately capture spatial dependency and temporal dependency and fail to capture the correlations between Different sEnsors at Different Timestamps (DEDT). Overlooking such correlations hinders the comprehensive modelling of ST dependencies within MTS data, thus restricting existing GNNs from learning effective representations. To address this limitation, we propose a novel method called Fully-Connected Spatial-Temporal Graph Neural Network (FC-STGNN), including two key components namely FC graph construction and FC graph convolution. For graph construction, we design a decay graph to connect sensors across all timestamps based on their temporal distances, enabling us to fully model the ST dependencies by considering the correlations between DEDT. Further, we devise FC graph convolution with a moving-pooling GNN layer to effectively capture the ST dependencies for learning effective representations. Extensive experiments show the effectiveness of FC-STGNN on multiple MTS datasets compared to SOTA methods. The code is available at https://github.com/Frank-Wang-oss/FCSTGNN.

PDF Details DOI

AAAI Conference 2024 Conference Paper

Graph-Aware Contrasting for Multivariate Time-Series Classification

Yucheng Wang
Yuecong Xu
Jianfei Yang
Min Wu
Xiaoli Li
Lihua Xie
Zhenghua Chen

Contrastive learning, as a self-supervised learning paradigm, becomes popular for Multivariate Time-Series (MTS) classification. It ensures the consistency across different views of unlabeled samples and then learns effective representations for these samples. Existing contrastive learning methods mainly focus on achieving temporal consistency with temporal augmentation and contrasting techniques, aiming to preserve temporal patterns against perturbations for MTS data. However, they overlook spatial consistency that requires the stability of individual sensors and their correlations. As MTS data typically originate from multiple sensors, ensuring spatial consistency becomes essential for the overall performance of contrastive learning on MTS data. Thus, we propose Graph-Aware Contrasting for spatial consistency across MTS data. Specifically, we propose graph augmentations including node and edge augmentations to preserve the stability of sensors and their correlations, followed by graph contrasting with both node- and graph-level contrasting to extract robust sensor- and global-level features. We further introduce multi-window temporal contrasting to ensure temporal consistency in the data for each sensor. Extensive experiments demonstrate that our proposed method achieves state-of-the-art performance on various MTS classification tasks. The code is available at https://github.com/Frank-Wang-oss/TS-GAC.

PDF Details DOI

TMLR Journal 2024 Journal Article

Hashing with Uncertainty Quantification via Sampling-based Hypothesis Testing

Yucheng Wang
Mingyuan Zhou
Xiaoning Qian

To quantify different types of uncertainty when deriving hash-codes for image retrieval, we develop a probabilistic hashing model(ProbHash). Sampling-based hypothesis testing is then derived for hashing with uncertainty quantification(HashUQ) in ProbHash to improve the granularity of hashing-based retrieval by prioritizing the data with confident hash-codes. HashUQ can drastically improve the retrieval performance without sacrificing computational efficiency. For efficient deployment of HashUQ in real-world applications, we discretize the quantified uncertainty to reduce the potential storage overhead. Experimental results show that our HashUQ can achieve state-of-the-art retrieval performance on three image datasets. Ablation experiments on model hyperparameters, different model components, and effects of UQ are also provided with performance comparisons. Our code is available at https://github.com/QianLab/HashUQ.

PDF Details

AIIM Journal 2023 Journal Article

Osteoporosis prediction in lumbar spine X-ray images using the multi-scale weighted fusion contextual transformer network

Linyan Xue
Geng Qin
Shilong Chang
Cheng Luo
Ya Hou
Zhiyin Xia
Jiacheng Yuan
Yucheng Wang

Osteoporosis is a bone-related disease characterized by decreased bone density and mass, leading to brittle fractures. Osteoporosis assessment from radiographs using a deep learning algorithm has proven a low-cost alternative to the golden standard DXA. Due to the considerable noise and low contrast, automated diagnosis of osteoporosis in X-ray images still poses a significant challenge for traditional diagnostic methods. In this paper, an end-to-end transformer-style network was proposed, termed FCoTNet, to overcome the shortcoming of insufficient fusion of texture information and local features in the traditional CoTNet. To extract complementary geometric representations at each scale of the transformer module, we integrated parallel multi-scale feature extraction architectures in each unit layer of FCoTNet to utilize convolution to aggregate features from different receptive fields. Moreover, in order to extract small-scale texture features which were more critical to the diagnosis of osteoporosis in radiographs, larger fusion weights were assigned to the feature maps with small-size receptive fields. Afterward, the multi-scale global modeling was conducted by self-attention mechanism. The proposed model was first investigated on a private lumbar spine X-ray dataset with the 5-fold cross-validation strategy, obtaining an average accuracy of 78. 29 ± 0. 93 %, an average sensitivity of 69. 72 ± 2. 35 %, and an average specificity of 88. 92 ± 0. 67 % for the multi-classification of normal, osteopenia, and osteoporosis categories. We then conducted a controlled trial with five orthopedic clinicians to evaluate the clinical value of the model. The average clinician's accuracy improved from 61. 50 ± 10. 79 % unaided to 80. 00 ± 5. 92 % aided (18. 50 % improvement), sensitivity improved from 64. 38 ± 8. 07 % unaided to 83. 31 ± 5. 43 % aided (18. 93 % improvement), and specificity improved from 80. 11 ± 4. 72 % unaided to 89. 94 ± 3. 82 % aided (9. 83 % improvement). Meanwhile, the prediction consistency among clinicians significantly improved with the assistance of FCoTNet. Furthermore, the proposed model showed good robustness on an external test dataset. These investigations indicate that the proposed deep learning model achieves state-of-the-art performance for osteoporosis prediction, which substantially improves osteoporosis screening and reduced osteoporosis fractures.

Details DOI

AAAI Conference 2023 Conference Paper

SEnsor Alignment for Multivariate Time-Series Unsupervised Domain Adaptation

Yucheng Wang
Yuecong Xu
Jianfei Yang
Zhenghua Chen
Min Wu
Xiaoli Li
Lihua Xie

Unsupervised Domain Adaptation (UDA) methods can reduce label dependency by mitigating the feature discrepancy between labeled samples in a source domain and unlabeled samples in a similar yet shifted target domain. Though achieving good performance, these methods are inapplicable for Multivariate Time-Series (MTS) data. MTS data are collected from multiple sensors, each of which follows various distributions. However, most UDA methods solely focus on aligning global features but cannot consider the distinct distributions of each sensor. To cope with such concerns, a practical domain adaptation scenario is formulated as Multivariate Time-Series Unsupervised Domain Adaptation (MTS-UDA). In this paper, we propose SEnsor Alignment (SEA) for MTS-UDA to reduce the domain discrepancy at both the local and global sensor levels. At the local sensor level, we design the endo-feature alignment to align sensor features and their correlations across domains, whose information represents the features of each sensor and the interactions between sensors. Further, to reduce domain discrepancy at the global sensor level, we design the exo-feature alignment to enforce restrictions on the global sensor features. Meanwhile, MTS also incorporates the essential spatial-temporal dependencies information between sensors, which cannot be transferred by existing UDA methods. Therefore, we model the spatial-temporal information of MTS with a multi-branch self-attention mechanism for simple and effective transfer across domains. Empirical results demonstrate the state-of-the-art performance of our proposed SEA on two public MTS datasets for MTS-UDA. The code is available at https://github.com/Frank-Wang-oss/SEA

PDF Details DOI

TCS Journal 2019 Journal Article

Combinatorial properties of Farey graphs

Yucheng Wang
Qi Bao
Zhongzhi Zhang

Combinatorial problems are a fundamental research subject of theoretical computer science, and for a general graph many combinatorial problems are NP-hard and even #P-complete. Thus, it is interesting to seek or design special graphs for which these difficult combinatorial problems can be exactly solved. In this paper, we study some combinatorial problems for the Farey graphs, which are translated from Farey sequences and have received considerable attention from the scientific community. We determine exactly the domination number, the independence number, and the matching number. Moreover, we derive exact or recursive solutions to the number of minimum dominating sets, the number of dominating sets, the number of maximum independent sets, the number of independent sets, the number of maximum matchings, as well as the number of matchings. Finally, we obtain explicit expressions for the number of acyclic orientations and the number of root-connected acyclic orientations. Since the considered combinatorial problems have found wide applications in diverse fields, such as network science and graph data miming, this work is helpful for deepening our understanding of the applications for these combinatorial problems.

Details DOI

AAAI Conference 2018 Conference Paper

Deep Stereo Matching With Explicit Cost Aggregation Sub-Architecture

Lidong Yu
Yucheng Wang
Yuwei Wu
Yunde Jia

Deep neural networks have shown excellent performance for stereo matching. Many efforts focus on the feature extraction and similarity measurement of the matching cost computation step while less attention is paid on cost aggregation which is crucial for stereo matching. In this paper, we present a learning-based cost aggregation method for stereo matching by a novel sub-architecture in the end-to-end trainable pipeline. We reformulate the cost aggregation as a learning process of the generation and selection of cost aggregation proposals which indicate the possible cost aggregation results. The cost aggregation sub-architecture is realized by a two-stream network: one for the generation of cost aggregation proposals, the other for the selection of the proposals. The criterion for the selection is determined by the low-level structure information obtained from a light convolutional network. The two-stream network offers a global view guidance for the cost aggregation to rectify the mismatching value stemming from the limited view of the matching cost computation. The comprehensive experiments on challenge datasets such as KITTI and Scene Flow show that our method outperforms the state-of-the-art methods.

PDF Details