PW
Pengxu Wei
cs.CVcs.AIcs.CLeess.IVcs.LGAuxiliary diagnosisCapsule endoscopyCrohn’s diseasecs.MMDeep Learning
On Valency
published · living versionsW_kjexj72q·v1 · currentpublished
Memory-Centric Continual Learning for AI Agents
with Dongfang Li, Zhuoen Chen, Bonian Jia, Zixuan Liu +23
1 version
Preprints & journals
42 papers in the corpus · 2017–2026Memory-Centric Continual Learning for AI Agents2610.29357v1 · Dongfang Li, Zhuoen Chen, Bonian Jia et al.2026 · 0 citationsarXivon Valency
The Right Lesson at the Right Step: Deriving Control Updates for Self-Evolving Agents2609.34988v1 · Yunhe Su, ZiYi Dong, Tong Yu et al.2026 · 0 citationsarXiv
EvoSCM: Scientific Belief Revision Through Causal Model Evolution and Experimentation2609.01526v2 · Qing Zhao, Haowei Li, Weijian Deng et al.2026 · 0 citationsarXiv
Ground4D: Consistency-Aware 4D Reconstruction from Monocular Video2606.28828v2 · Qing Zhao, Weijian Deng, Pengxu Wei et al.2026 · 0 citationsarXiv
Where Detectors Fail: Probing Generative Space for Generalizable AI-Generated Image Detection2605.24906v3 · Zijie Cao, Weijie Tu, Yao Xiao et al.2026 · 0 citationsarXiv
When Preference Labels Fall Short: Aligning Diffusion Models from Real Data2605.19839v2 · Weiyan Chen, Weijian Deng, Yao Xiao et al.2026 · 0 citationsarXiv
Language Generation as Optimal Control: Closed-Loop Diffusion in Latent Control Space2605.14531v3 · ZiYi Dong, Yuliang Huang, Weijian Deng et al.2026 · 0 citationsarXiv
Unveiling Perceptual Artifacts: A Fine-Grained Benchmark for Interpretable AI-Generated Image Detection2601.19430v1 · Yao Xiao, Weiyan Chen, Jiahao Chen et al.2026 · 0 citationsarXiv
Robust Source-Free Domain Adaptation From Non-Robust Source Models.41719580 · Xiao, Yao, Wei, Pengxu, Wang, Guangrun et al.2026 · 0 citationsIEEE transactions on image processing : a publication of the IEEE Signal Processing Society. 2026;35:2350-2363
Delving into Cascaded Instability: A Lipschitz Continuity View on Image Restoration and Object Detection Synergy2510.24232v3 · Qing Zhao, Weijian Deng, Pengxu Wei et al.2025 · 2 citationsarXiv
Can We Achieve Efficient Diffusion without Self-Attention? Distilling Self-Attention into Convolutions2504.21292v1 · ZiYi Dong, Chengxing Zhou, Weijian Deng et al.2025 · 0 citationsarXiv
Decoder-Only LLMs are Better Controllers for Diffusion Models2502.04412v1 · Ziyi Dong, Yao Xiao, Pengxu Wei et al.2025 · 3 citationsarXiv
CIPL: Counterfactual Interactive Policy Learning to Eliminate Popularity Bias for Online Recommendation.37585330 · Zheng, Yongsen, Qin, Jinghui, Wei, Pengxu et al.2025 · 6 citationsIEEE transactions on neural networks and learning systems. 2024;35(12):17123-17136
Routing User-Interest Markov Tree for Scalable Personalized Knowledge-Aware Recommendation.37247310 · Zheng, Yongsen, Wei, Pengxu, Chen, Ziliang et al.2024 · 1 citationIEEE transactions on neural networks and learning systems. 2024;35(10):14233-14246
SAM-COD: SAM-guided Unified Framework for Weakly-Supervised Camouflaged Object Detection2408.10760v1 · Huafeng Chen, Pengxu Wei, Guangqian Guo et al.2024 · 31 citationsarXiv
Diff-Mosaic: Augmenting Realistic Representations in Infrared Small Target Detection via Diffusion Prior2406.00632v1 · Yukai Shi, Yupei Lin, Pengxu Wei et al.2024 · 13 citationsarXiv
Towards Understanding the Robustness of Diffusion-Based Purification: A Stochastic Perspective2404.14309v3 · Yiming Liu, Kezhao Liu, Yao Xiao et al.2024 · 0 citationsarXiv
NTIRE 2024 Challenge on Image Super-Resolution (x4): Methods and Results2404.09790v2 · Zheng Chen, Zongwei Wu, Eduard Zamfir et al.2024 · 56 citationsProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2024, pp. 6108-6132
IDF-CR: Iterative Diffusion Process for Divide-and-Conquer Cloud Removal in Remote-sensing Images2403.11870v1 · Meilin Wang, Yexing Song, Pengxu Wei et al.2024 · 47 citationsarXiv
Image Restoration Through Generalized Ornstein-Uhlenbeck Bridge2312.10299v2 · Conghan Yue, Zhengwei Peng, Junlong Ma et al.2023 · 7 citationsarXiv
Towards Real-World Burst Image Super-Resolution: Benchmark and Method2309.04803v1 · Pengxu Wei, Yujing Sun, Xingbei Guo et al.2023 · 27 citationsarXiv
Identity-Preserving Talking Face Generation with Landmark and Appearance Priors2305.08293v1 · Weizhi Zhong, Chaowei Fang, Yinqi Cai et al.2023 · 103 citationsarXiv
Masked Images Are Counterfactual Samples for Robust Fine-tuning2303.03052v3 · Yao Xiao, Ziyi Tang, Pengxu Wei et al.2023 · 18 citationsarXiv
Taylor Neural Network for Real-World Image Super-Resolution.37028347 · Wei, Pengxu, Xie, Ziwei, Li, Guanbin et al.2025 · 19 citationsIEEE transactions on image processing : a publication of the IEEE Signal Processing Society. 2023;32:1942-1951
DQnet: Cross-Model Detail Querying for Camouflaged Object Detection2212.08296v1 · Wei Sun, Chengao Liu, Linyan Zhang et al.2022 · 3 citationsarXiv
Out-of-Candidate Rectification for Weakly Supervised Semantic Segmentation2211.12268v3 · Zesen Cheng, Pengchong Qiao, Kehan Li et al.2022 · 78 citationsarXiv
Robust Real-World Image Super-Resolution against Adversarial Attacks2208.00428v1 · Jiutao Yue, Haofeng Li, Pengxu Wei et al.2022 · 20 citationsProceedings of the 29th ACM International Conference on Multimedia (2021) 5148-5157
Adversarially-Aware Robust Object Detector2207.06202v3 · Ziyi Dong, Pengxu Wei, Liang Lin2022 · 32 citationsEuropean Conference on Computer Vision (ECCV), 2022
Real-World Image Super-Resolution by Exclusionary Dual-Learning2206.02609v1 · Hao Li, Jinghui Qin, Zhijing Yang et al.2022 · 25 citationsarXiv
Dual Adversarial Adaptation for Cross-Device Real-World Image Super-Resolution2205.03524v1 · Xiaoqian Xu, Pengxu Wei, Weikai Chen et al.2022 · 20 citationsarXiv
NTIRE 2022 Challenge on Super-Resolution and Quality Enhancement of Compressed Video: Dataset, Methods and Results2204.09314v2 · Ren Yang, Radu Timofte, Meisong Zheng et al.2022 · 31 citationsarXiv
Open Set Domain Adaptation By Novel Class Discovery2203.03329v1 · Jingyu Zhuang, Ziliang Chen, Pengxu Wei et al.2022 · 9 citationsarXiv
Deductive Reinforcement Learning for Visual Autonomous Urban Driving Navigation.34520367 · Huang, Changxin, Zhang, Ronghui, Ouyang, Meizi et al.2021 · 50 citationsIEEE transactions on neural networks and learning systems. 2021;32(12):5379-5391
Multi-Task Classification and Segmentation for Explicable Capsule Endoscopy Diagnostics.34490342 · Kong, Zishang, He, Min, Luo, Qianjiang et al.2024 · 18 citationsFrontiers in molecular biosciences. 2021;8:614277
Trash to Treasure: Harvesting OOD Data with Cross-Modal Matching for Open-Set Semi-Supervised Learning2108.05617v1 · Junkai Huang, Chaowei Fang, Weikai Chen et al.2021 · 49 citationsarXiv
AIM 2020 Challenge on Real Image Super-Resolution: Methods and Results2009.12072v1 · Pengxu Wei, Hannan Lu, Radu Timofte et al.2020 · 37 citationsEuropean Conference on Computer Vision Workshops, 2020
Component Divide-and-Conquer for Real-World Image Super-Resolution2008.01928v1 · Pengxu Wei, Ziwei Xie, Hannan Lu et al.2020 · 278 citationsEuropean Conference on Computer Vision (ECCV), 2020
A Near-Optimal Gradient Flow for Learning Neural Energy-Based Models1910.14216v7 · Yang Wu, Pengxu Wei, Liang Lin2019 · 0 citationsarXiv
Min-Entropy Latent Model for Weakly Supervised Object Detection1902.06057v1 · Fang Wan, Pengxu Wei, Zhenjun Han et al.2019 · 201 citationsarXiv
Min-Entropy Latent Model for Weakly Supervised Object Detection.30762529 · Wan, Fang, Wei, Pengxu, Han, Zhenjun et al.2019 · 92 citationsIEEE transactions on pattern analysis and machine intelligence. 2019;41(10):2395-2409
3D Human Pose Machines with Self-Supervised Learning.30640601 · Wang, Keze, Lin, Liang, Jiang, Chenhan et al.2021 · 97 citationsIEEE transactions on pattern analysis and machine intelligence. 2020;42(5):1069-1082
Correlated Topic Vector for Scene Classification.28422661 · Wei, Pengxu, Qin, Fei, Wan, Fang et al.2018 · 3 citationsIEEE transactions on image processing : a publication of the IEEE Signal Processing Society. 2017;26(7):3221-3234
Career total: 85 works. 42 are in this corpus.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.