Search

Your search keyword '"Fan, Qi"' showing total 3,817 results
Start Over Author "Fan, Qi"
3,817 results on '"Fan, Qi"'

1. Domain-Rectifying Adapter for Cross-Domain Few-Shot Segmentation

Author

Su, Jiapeng, Fan, Qi, Lu, Guangming, Chen, Fanglin, and Pei, Wenjie

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Few-shot semantic segmentation (FSS) has achieved great success on segmenting objects of novel classes, supported by only a few annotated samples. However, existing FSS methods often underperform in the presence of domain shifts, especially when encountering new domain styles that are unseen during training. It is suboptimal to directly adapt or generalize the entire model to new domains in the few-shot scenario. Instead, our key idea is to adapt a small adapter for rectifying diverse target domain styles to the source domain. Consequently, the rectified target domain features can fittingly benefit from the well-optimized source domain segmentation model, which is intently trained on sufficient source domain data. Training domain-rectifying adapter requires sufficiently diverse target domains. We thus propose a novel local-global style perturbation method to simulate diverse potential target domains by perturbating the feature channel statistics of the individual images and collective statistics of the entire source domain, respectively. Additionally, we propose a cyclic domain alignment module to facilitate the adapter effectively rectifying domains using a reverse domain rectification supervision. The adapter is trained to rectify the image features from diverse synthesized target domains to align with the source domain. During testing on target domains, we start by rectifying the image features and then conduct few-shot segmentation on the domain-rectified features. Extensive experiments demonstrate the effectiveness of our method, achieving promising results on cross-domain few-shot semantic segmentation tasks. Our code is available at https://github.com/Matt-Su/DR-Adapter., Comment: Accepted by CVPR 2024

Published
2024

2. ONNXPruner: ONNX-Based General Model Pruning Adapter

Author

Ren, Dongdong, Li, Wenbin, Ding, Tianyu, Wang, Lei, Fan, Qi, Huo, Jing, Pan, Hongbing, and Gao, Yang

Subjects
Computer Science - Machine Learning
Abstract

Recent advancements in model pruning have focused on developing new algorithms and improving upon benchmarks. However, the practical application of these algorithms across various models and platforms remains a significant challenge. To address this challenge, we propose ONNXPruner, a versatile pruning adapter designed for the ONNX format models. ONNXPruner streamlines the adaptation process across diverse deep learning frameworks and hardware platforms. A novel aspect of ONNXPruner is its use of node association trees, which automatically adapt to various model architectures. These trees clarify the structural relationships between nodes, guiding the pruning process, particularly highlighting the impact on interconnected nodes. Furthermore, we introduce a tree-level evaluation method. By leveraging node association trees, this method allows for a comprehensive analysis beyond traditional single-node evaluations, enhancing pruning performance without the need for extra operations. Experiments across multiple models and datasets confirm ONNXPruner's strong adaptability and increased efficacy. Our work aims to advance the practical application of model pruning.

Published
2024

3. InternLM2 Technical Report

Author

Cai, Zheng, Cao, Maosong, Chen, Haojiong, Chen, Kai, Chen, Keyu, Chen, Xin, Chen, Xun, Chen, Zehui, Chen, Zhi, Chu, Pei, Dong, Xiaoyi, Duan, Haodong, Fan, Qi, Fei, Zhaoye, Gao, Yang, Ge, Jiaye, Gu, Chenya, Gu, Yuzhe, Gui, Tao, Guo, Aijia, Guo, Qipeng, He, Conghui, Hu, Yingfan, Huang, Ting, Jiang, Tao, Jiao, Penglong, Jin, Zhenjiang, Lei, Zhikai, Li, Jiaxing, Li, Jingwen, Li, Linyang, Li, Shuaibin, Li, Wei, Li, Yining, Liu, Hongwei, Liu, Jiangning, Hong, Jiawei, Liu, Kaiwen, Liu, Kuikun, Liu, Xiaoran, Lv, Chengqi, Lv, Haijun, Lv, Kai, Ma, Li, Ma, Runyuan, Ma, Zerun, Ning, Wenchang, Ouyang, Linke, Qiu, Jiantao, Qu, Yuan, Shang, Fukai, Shao, Yunfan, Song, Demin, Song, Zifan, Sui, Zhihao, Sun, Peng, Sun, Yu, Tang, Huanze, Wang, Bin, Wang, Guoteng, Wang, Jiaqi, Wang, Jiayu, Wang, Rui, Wang, Yudong, Wang, Ziyi, Wei, Xingjian, Weng, Qizhen, Wu, Fan, Xiong, Yingtong, Xu, Chao, Xu, Ruiliang, Yan, Hang, Yan, Yirong, Yang, Xiaogui, Ye, Haochen, Ying, Huaiyuan, Yu, Jia, Yu, Jing, Zang, Yuhang, Zhang, Chuyu, Zhang, Li, Zhang, Pan, Zhang, Peng, Zhang, Ruijie, Zhang, Shuo, Zhang, Songyang, Zhang, Wenjian, Zhang, Wenwei, Zhang, Xingcheng, Zhang, Xinyue, Zhao, Hui, Zhao, Qian, Zhao, Xiaomeng, Zhou, Fengzhe, Zhou, Zaida, Zhuo, Jingming, Zou, Yicheng, Qiu, Xipeng, Qiao, Yu, and Lin, Dahua

Subjects
Computer Science - Computation and Language, Computer Science - Artificial Intelligence
Abstract

The evolution of Large Language Models (LLMs) like ChatGPT and GPT-4 has sparked discussions on the advent of Artificial General Intelligence (AGI). However, replicating such advancements in open-source models has been challenging. This paper introduces InternLM2, an open-source LLM that outperforms its predecessors in comprehensive evaluations across 6 dimensions and 30 benchmarks, long-context modeling, and open-ended subjective evaluations through innovative pre-training and optimization techniques. The pre-training process of InternLM2 is meticulously detailed, highlighting the preparation of diverse data types including text, code, and long-context data. InternLM2 efficiently captures long-term dependencies, initially trained on 4k tokens before advancing to 32k tokens in pre-training and fine-tuning stages, exhibiting remarkable performance on the 200k ``Needle-in-a-Haystack" test. InternLM2 is further aligned using Supervised Fine-Tuning (SFT) and a novel Conditional Online Reinforcement Learning from Human Feedback (COOL RLHF) strategy that addresses conflicting human preferences and reward hacking. By releasing InternLM2 models in different training stages and model sizes, we provide the community with insights into the model's evolution.

Published
2024

4. A Spark Optimizer for Adaptive, Fine-Grained Parameter Tuning

Author

Lyu, Chenghao, Fan, Qi, Guyard, Philippe, and Diao, Yanlei

Subjects
Computer Science - Distributed, Parallel, and Cluster Computing
Abstract

As Spark becomes a common big data analytics platform, its growing complexity makes automatic tuning of numerous parameters critical for performance. Our work on Spark parameter tuning is particularly motivated by two recent trends: Spark's Adaptive Query Execution (AQE) based on runtime statistics, and the increasingly popular Spark cloud deployments that make cost-performance reasoning crucial for the end user. This paper presents our design of a Spark optimizer that controls all tunable parameters of each query in the new AQE architecture to explore its performance benefits and, at the same time, casts the tuning problem in the theoretically sound multi-objective optimization (MOO) setting to better adapt to user cost-performance preferences. To this end, we propose a novel hybrid compile-time/runtime approach to multi-granularity tuning of diverse, correlated Spark parameters, as well as a suite of modeling and optimization techniques to solve the tuning problem in the MOO setting while meeting the stringent time constraint of 1-2 seconds for cloud use. Evaluation results using TPC-H and TPC-DS benchmarks demonstrate the superior performance of our approach: (i) When prioritizing latency, it achieves 63% and 65% reduction for TPC-H and TPC-DS, respectively, under an average solving time of 0.7-0.8 sec, outperforming the most competitive MOO method that reduces only 18-25% latency with 2.6-15 sec solving time. (ii) When shifting preferences between latency and cost, our approach dominates the solutions of alternative methods, exhibiting superior adaptability to varying preferences.

Published
2024

7. DARNet: Bridging Domain Gaps in Cross-Domain Few-Shot Segmentation with Dynamic Adaptation

Author

Fan, Haoran, Fan, Qi, Pagnucco, Maurice, and Song, Yang

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Few-shot segmentation (FSS) aims to segment novel classes in a query image by using only a small number of supporting images from base classes. However, in cross-domain few-shot segmentation (CD-FSS), leveraging features from label-rich domains for resource-constrained domains poses challenges due to domain discrepancies. This work presents a Dynamically Adaptive Refine (DARNet) method that aims to balance generalization and specificity for CD-FSS. Our method includes the Channel Statistics Disruption (CSD) strategy, which perturbs feature channel statistics in the source domain, bolstering generalization to unknown target domains. Moreover, recognizing the variability across target domains, an Adaptive Refine Self-Matching (ARSM) method is also proposed to adjust the matching threshold and dynamically refine the prediction result with the self-matching method, enhancing accuracy. We also present a Test-Time Adaptation (TTA) method to refine the model's adaptability to diverse feature distributions. Our approach demonstrates superior performance against state-of-the-art methods in CD-FSS tasks.

Published
2023

8. Stable Segment Anything Model

Author

Fan, Qi, Tao, Xin, Ke, Lei, Ye, Mingqiao, Zhang, Yuan, Wan, Pengfei, Wang, Zhongyuan, Tai, Yu-Wing, and Tang, Chi-Keung

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

The Segment Anything Model (SAM) achieves remarkable promptable segmentation given high-quality prompts which, however, often require good skills to specify. To make SAM robust to casual prompts, this paper presents the first comprehensive analysis on SAM's segmentation stability across a diverse spectrum of prompt qualities, notably imprecise bounding boxes and insufficient points. Our key finding reveals that given such low-quality prompts, SAM's mask decoder tends to activate image features that are biased towards the background or confined to specific object parts. To mitigate this issue, our key idea consists of calibrating solely SAM's mask attention by adjusting the sampling locations and amplitudes of image features, while the original SAM model architecture and weights remain unchanged. Consequently, our deformable sampling plugin (DSP) enables SAM to adaptively shift attention to the prompted target regions in a data-driven manner, facilitated by our effective robust training strategy (RTS). During inference, dynamic routing plugin (DRP) is proposed that toggles SAM between the deformable and regular grid sampling modes, conditioned on the input prompt quality. Thus, our solution, termed Stable-SAM, offers several advantages: 1) improved SAM's segmentation stability across a wide range of prompt qualities, while 2) retaining SAM's powerful promptable segmentation efficiency and generality, with 3) minimal learnable parameters (0.08 M) and fast adaptation (by 1 training epoch). Extensive experiments across multiple datasets validate the effectiveness and advantages of our approach, underscoring Stable-SAM as a more robust solution for segmenting anything. Codes will be released upon acceptance. https://github.com/fanq15/Stable-SAM, Comment: Smaller file size for the easy access. Codes will be released upon acceptance. https://github.com/fanq15/Stable-SAM

Published
2023

9. Selective Feature Adapter for Dense Vision Transformers

Author

Deng, Xueqing, Fan, Qi, Jin, Xiaojie, Yang, Linjie, and Wang, Peng

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Fine-tuning pre-trained transformer models, e.g., Swin Transformer, are successful in numerous downstream for dense prediction vision tasks. However, one major issue is the cost/storage of their huge amount of parameters, which becomes increasingly challenging to handle with the growing amount of vision tasks. In this paper, we propose an effective approach to alleviate the issue, namely selective feature adapter (SFA). It achieves state-of-the-art (SoTA) performance under any given budget of trainable parameters, and demonstrates comparable or better performance than fully fine-tuned models across various dense tasks. Specifically, SFA consists of external adapters and internal adapters which are sequentially operated over a transformer model. For external adapters, we properly select the places and amount of additional multilayer perception (MLP). For internal adapters, we transform a few task-important parameters inside the transformer, which are automatically discovered through a simple yet effective lottery ticket algorithm. Our experiments show that the dual adapter module, a.k.a SFA, is essential to achieve the best trade-off on dense vision tasks, such as segmentation, detection and depth-estimation, outperforming other adapters with a single module.

Published
2023

11. Learning Noise-Robust Joint Representation for Multimodal Emotion Recognition under Incomplete Data Scenarios

Author

Fan, Qi, Zuo, Haolin, Liu, Rui, Lian, Zheng, and Gao, Guanglai

Subjects
Computer Science - Computer Vision and Pattern Recognition, Computer Science - Artificial Intelligence, Computer Science - Machine Learning
Abstract

Multimodal emotion recognition (MER) in practical scenarios is significantly challenged by the presence of missing or incomplete data across different modalities. To overcome these challenges, researchers have aimed to simulate incomplete conditions during the training phase to enhance the system's overall robustness. Traditional methods have often involved discarding data or substituting data segments with zero vectors to approximate these incompletenesses. However, such approaches neither accurately represent real-world conditions nor adequately address the issue of noisy data availability. For instance, a blurry image cannot be simply replaced with zero vectors, and still retain information. To tackle this issue and develop a more precise MER system, we introduce a novel noise-robust MER model that effectively learns robust multimodal joint representations from noisy data. This approach includes two pivotal components: firstly, a noise scheduler that adjusts the type and level of noise in the data to emulate various realistic incomplete situations. Secondly, a Variational AutoEncoder (VAE)-based module is employed to reconstruct these robust multimodal joint representations from the noisy inputs. Notably, the introduction of the noise scheduler enables the exploration of an entirely new type of incomplete data condition, which is impossible with existing methods. Extensive experimental evaluations on the benchmark datasets IEMOCAP and CMU-MOSEI demonstrate the effectiveness of the noise scheduler and the excellent performance of our proposed model.

Published
2023
Full Text
View/download PDF

13. Synthetic lethality of combined ULK1 defection and p53 restoration induce pyroptosis by directly upregulating GSDME transcription and cleavage activation through ROS/NLRP3 signaling

Author

Wei Chen, Kai-Bin Yang, Yuan-Zhe Zhang, Zai-Shan Lin, Jin-Wei Chen, Si-Fan Qi, Chen-Fei Wu, Gong-Kan Feng, Da-Jun Yang, Ming Chen, Xiao-Feng Zhu, and Xuan Li

Subjects
MDM2 inhibitor, Mitophagy, Pyroptosis, Reactive oxygen species, TP53, ULK1, Neoplasms. Tumors. Oncology. Including cancer and carcinogens, RC254-282
Abstract

Abstract Background High expression of ubiquitin ligase MDM2 is a primary cause of p53 inactivation in many tumors, making it a promising therapeutic target. However, MDM2 inhibitors have failed in clinical trials due to p53-induced feedback that enhances MDM2 expression. This underscores the urgent need to find an effective adaptive genotype or combination of targets. Methods Kinome-wide CRISPR/Cas9 knockout screen was performed to identify genes that modulate the response to MDM2 inhibitor using TP53 wild type cancer cells and found ULK1 as a candidate. The MTT cell viability assay, flow cytometry and LDH assay were conducted to evaluate the activation of pyroptosis and the synthetic lethality effects of combining ULK1 depletion with p53 activation. Dual-luciferase reporter assay and ChIP-qPCR were performed to confirm that p53 directly mediates the transcription of GSDME and to identify the binding region of p53 in the promoter of GSDME. ULK1 knockout / overexpression cells were constructed to investigate the functional role of ULK1 both in vitro and in vivo. The mechanism of ULK1 depletion to activate GSMDE was mainly investigated by qPCR, western blot and ELISA. Results By using high-throughput screening, we identified ULK1 as a synthetic lethal gene for the MDM2 inhibitor APG115. It was determined that deletion of ULK1 significantly increased the sensitivity, with cells undergoing typical pyroptosis. Mechanistically, p53 promote pyroptosis initiation by directly mediating GSDME transcription that induce basal-level pyroptosis. Moreover, ULK1 depletion reduces mitophagy, resulting in the accumulation of damaged mitochondria and subsequent increasing of reactive oxygen species (ROS). This in turn cleaves and activates GSDME via the NLRP3-Caspase inflammatory signaling axis. The molecular cascade makes ULK1 act as a crucial regulator of pyroptosis initiation mediated by p53 activation cells. Besides, mitophagy is enhanced in platinum-resistant tumors, and ULK1 depletion/p53 activation has a synergistic lethal effect on these tumors, inducing pyroptosis through GSDME directly. Conclusion Our research demonstrates that ULK1 deficiency can synergize with MDM2 inhibitors to induce pyroptosis. p53 plays a direct role in activating GSDME transcription, while ULK1 deficiency triggers upregulation of the ROS-NLRP3 signaling pathway, leading to GSDME cleavage and activation. These findings underscore the pivotal role of p53 in determining pyroptosis and provide new avenues for the clinical application of p53 restoration therapies, as well as suggesting potential combination strategies.

Published
2024
Full Text
View/download PDF

14. Automatic Measurement Method of Beef Cattle Body Size Based on Multimodal Image Information and Improved Instance Segmentation Network

Author

WENG Zhi, FAN Qi, and ZHENG Zhiqiang

Subjects
cattle body size measurement, deep learning, point cloud segmentation, instance segmentation, attention mechanism, mask2former, Agriculture (General), S1-972, Technology (General), T1-995
Abstract

ObjectiveThe body size parameter of cattle is a key indicator reflecting the physical development of cattle, and is also a key factor in the cattle selection and breeding process. In order to solve the demand of measuring body size of beef cattle in the complex environment of large-scale beef cattle ranch, an image acquisition device and an automatic measurement algorithm of body size were designed.MethodsFirstly, the walking channel of the beef cattle was established, and when the beef cattle entered the restraining device through the channel, the RGB and depth maps of the image on the right side of the beef cattle were acquired using the Inter RealSense D455 camera. Secondly, in order to avoid the influence of the complex environmental background, an improved instance segmentation network based on Mask2former was proposed, adding CBAM module and CA module, respectively, to improve the model's ability to extract key features from different perspectives, extracting the foreground contour from the 2D image of the cattle, partitioning the contour, and comparing it with other segmentation algorithms, and using curvature calculation and other mathematical methods to find the required body size measurement points. Thirdly, in the processing of 3D data, in order to solve the problem that the pixel point to be measured in the 2D RGB image was null when it was projected to the corresponding pixel coordinates in the depth-valued image, resulting in the inability to calculate the 3D coordinates of the point, a series of processing was performed on the point cloud data, and a suitable point cloud filtering and point cloud segmentation algorithm was selected to effectively retain the point cloud data of the region of the cattle's body to be measured, and then the depth map was 16. Then the depth map was filled with nulls in the field to retain the integrity of the point cloud in the cattle body region, so that the required measurement points could be found and the 2D data could be returned. Finally, an extraction algorithm was designed to combine 2D and 3D data to project the extracted 2D pixel points into a 3D point cloud, and the camera parameters were used to calculate the world coordinates of the projected points, thus automatically calculating the body measurements of the beef cattle.Results and DiscussionsFirstly, in the part of instance segmentation, compared with the classical Mask R-CNN and the recent instance segmentation networks PointRend and Queryinst, the improved network could extract higher precision and smoother foreground images of cattles in terms of segmentation accuracy and segmentation effect, no matter it was for the case of occlusion or for the case of multiple cattles. Secondly, in three-dimensional data processing, the method proposed in the study could effectively extract the three-dimensional data of the target area. Thirdly, the measurement error of body size was analysed, among the four body size measurement parameters, the smallest average relative error was the height of the cross section, which was due to the more prominent position of the cross section, and the different standing positions of the cattle have less influence on the position of the cross section, and the largest average relative error was the pipe circumference, which was due to the influence of the greater overlap of the two front legs, and the higher requirements for the standing position. Finally, automatic body measurements were carried out on 137 beef cattle in the ranch, and the automatic measurements of the four body measurements parameters were compared with the manual measurements, and the results showed that the average relative errors of body height, cross section height, body slant length, and tube girth were 4.32%, 3.71%, 5.58% and 6.25%, respectively, which met the needs of the ranch. The shortcomings were that fewer body-size parameters were measured, and the error of measuring circumference-type body-size parameters was relatively large. Later studies could use a multi-view approach to increase the number of body rule parameters to be measured and improve the accuracy of the parameters in the circumference category.ConclusionsThe article designed an automatic measurement method based on two-dimensional and three-dimensional contactless body measurements of beef cattle. Moreover, the innovatively proposed method of measuring tube girth has higher accuracy and better implementation compared with the current research on body measurements in beef cattle. The relative average errors of the four body tape parameters meet the needs of pasture measurements and provide theoretical and practical guidance for the automatic measurement of body tape in beef cattle.

Published
2024
Full Text
View/download PDF

15. UniBoost: Unsupervised Unimodal Pre-training for Boosting Zero-shot Vision-Language Tasks

Author

Sun, Yanan, Zhong, Zihan, Fan, Qi, Tang, Chi-Keung, and Tai, Yu-Wing

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Large-scale joint training of multimodal models, e.g., CLIP, have demonstrated great performance in many vision-language tasks. However, image-text pairs for pre-training are restricted to the intersection of images and texts, limiting their ability to cover a large distribution of real-world data, where noise can also be introduced as misaligned pairs during pre-processing. Conversely, unimodal models trained on text or image data alone through unsupervised techniques can achieve broader coverage of diverse real-world data and are not constrained by the requirement of simultaneous presence of image and text. In this paper, we demonstrate that using large-scale unsupervised unimodal models as pre-training can enhance the zero-shot performance of image-text pair models. Our thorough studies validate that models pre-trained as such can learn rich representations of both modalities, improving their ability to understand how images and text relate to each other. Our experiments show that unimodal pre-training outperforms state-of-the-art CLIP-based models by 6.5% (52.3% $\rightarrow$ 58.8%) on PASCAL-5$^i$ and 6.2% (27.2% $\rightarrow$ 33.4%) on COCO-20$^i$ semantic segmentation under zero-shot setting respectively. By learning representations of both modalities, unimodal pre-training offers broader coverage, reduced misalignment errors, and the ability to capture more complex features and patterns in the real-world data resulting in better performance especially for zero-shot vision-language tasks.

Published
2023

16. Alpha-Ketoglutarate Regulates Tnfrsf12a/Fn14 Expression via Histone Modification and Prevents Cancer-Induced Cachexia.

Author

Ruiz, Bryan, Lowman, Xazmin, Yang, Ying, Fan, Qi, Wang, Tianhong, Wu, Hongmei, Hanse, Eric, and Kong, Mei

Subjects
FN14, TNFRSF12A, alpha-ketoglutarate, cachexia, cancer, colon cancer, histone, Animals, Mice, Cachexia, Ketoglutaric Acids, Histones, Glutamine, Histone Code, Colonic Neoplasms, Disease Models, Animal, Histone Methyltransferases, Tumor Microenvironment
Abstract

Previous studies have shown that inhibition of TNF family member FN14 (gene: TNFRSF12A) in colon tumors decreases inflammatory cytokine expression and mitigates cancer-induced cachexia. However, the molecular mechanisms underlying the regulation of FN14 expression remain unclear. Tumor microenvironments are often devoid of nutrients and oxygen, yet how the cachexic response relates to the tumor microenvironment and, importantly, nutrient stress is unknown. Here, we looked at the connections between metabolic stress and FN14 expression. We found that TNFRSF12A expression was transcriptionally induced during glutamine deprivation in cancer cell lines. We also show that the downstream glutaminolysis metabolite, alpha-ketoglutarate (aKG), is sufficient to rescue glutamine-deprivation-promoted TNFRSF12A induction. As aKG is a co-factor for histone de-methylase, we looked at histone methylation and found that histone H3K4me3 at the Tnfrsf12a promoter is increased under glutamine-deprived conditions and rescued via DM-aKG supplementation. Finally, expression of Tnfrsf12a and cachexia-induced weight loss can be inhibited in vivo by DM-aKG in a mouse cancer cachexia model. These findings highlight a connection between metabolic stress and cancer cachexia development.

Published
2023

17. Vitamin B6 is governed by the local compartmentalization of metabolic enzymes during growth.

Author

Franco, Carolina, Seabrook, Laurence, Nguyen, Steven, Yang, Ying, Campos, Melissa, Fan, Qi, Cicchetto, Andrew, Christofk, Heather, Albrecht, Lauren, and Kong, Mei

Subjects
Humans, Vitamin B 6, Proteolysis, Peptide Hydrolases, Micronutrients, Vitamins
Abstract

Vitamin B6 is a vital micronutrient across cell types and tissues, and dysregulated B6 levels contribute to human disease. Despite its importance, how B6 vitamer levels are regulated is not well understood. Here, we provide evidence that B6 dynamics are rapidly tuned by precise compartmentation of pyridoxal kinase (PDXK), the rate-limiting B6 enzyme. We show that canonical Wnt rapidly led to the accumulation of inactive B6 by shunting cytosolic PDXK into lysosomes. PDXK was modified with methyl-arginine Degron (MrDegron), a protein tag for lysosomes, which enabled delivery via microautophagy. Hyperactive lysosomes resulted in the continuous degradation of PDXK and B6 deficiency that promoted proliferation in Wnt-driven colorectal cancer (CRC) cells. Pharmacological or genetic disruption of the coordinated MrDegron proteolytic pathway was sufficient to reduce CRC survival in cells and organoid models. In sum, this work contributes to the repertoire of micronutrient-regulated processes that enable cancer cell growth and provides insight into the functional impact of B6 deficiencies for survival.

Published
2023

28. Stepped-up development of accelerator mass spectrometry method for the detection of 60Fe with the HI-13 tandem accelerator

Author

Zhang, Yang, Yan, Sheng-Quan, He, Ming, Zhao, Qing-Zhang, Zhang, Wen-Hui, Kan, Chao-Xin, Zhou, Jian-Ming, Li, Kang-Ning, Wang, Xiao-Fei, Liu, Jian-Cheng, Peng, Zhao-Hua, Liang, Zhuo, Li, Ai-Ling, Zheng, Jian, Fan, Qi-Wen, Li, Yun-Ju, Wang, You-Bao, Li, Zhi-Hong, Shen, Yang-Ping, Nan, Ding, Nan, Wei, Zhang, Yu-Qiang, Li, Jia-Ying-Hao, Tian, Jun-Wen, Hou, Jiang-Lin, Guo, Chang-Xin, Zhang, Zhi-Cheng, Zhu, Ming-Hao, Chen, Yu-Wen, Jiang, Yu-Chen, Tian, Tao, Ma, Jin-Long, Liu, Yi-Hui, Dong, Jing-Yu, Liu, Run-Long, Ma, Mei-Yue-Nan, Chen, Yong-Shou, Liu, Wei-Ping, and Guo, Bing

Published
2024
Full Text
View/download PDF

29. Normalization Perturbation: A Simple Domain Generalization Method for Real-World Domain Shifts

Author

Fan, Qi, Segu, Mattia, Tai, Yu-Wing, Yu, Fisher, Tang, Chi-Keung, Schiele, Bernt, and Dai, Dengxin

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Improving model's generalizability against domain shifts is crucial, especially for safety-critical applications such as autonomous driving. Real-world domain styles can vary substantially due to environment changes and sensor noises, but deep models only know the training domain style. Such domain style gap impedes model generalization on diverse real-world domains. Our proposed Normalization Perturbation (NP) can effectively overcome this domain style overfitting problem. We observe that this problem is mainly caused by the biased distribution of low-level features learned in shallow CNN layers. Thus, we propose to perturb the channel statistics of source domain features to synthesize various latent styles, so that the trained deep model can perceive diverse potential domains and generalizes well even without observations of target domain data in training. We further explore the style-sensitive channels for effective style synthesis. Normalization Perturbation only relies on a single source domain and is surprisingly effective and extremely easy to implement. Extensive experiments verify the effectiveness of our method for generalizing models under real-world domain shifts.

Published
2022

32. SAPS3 subunit of protein phosphatase 6 is an AMPK inhibitor and controls metabolic homeostasis upon dietary challenge in male mice

Author

Yang, Ying, Reid, Michael A, Hanse, Eric A, Li, Haiqing, Li, Yuanding, Ruiz, Bryan I, Fan, Qi, and Kong, Mei

Subjects
Biochemistry and Cell Biology, Biological Sciences, Genetics, Digestive Diseases, Nutrition, 2.1 Biological and endogenous factors, 1.1 Normal biological development and functioning, Underpinning research, Aetiology, Metabolic and endocrine, Animals, Male, Mice, AMP-Activated Protein Kinases, Diet, High-Fat, Homeostasis, Mice, Inbred C57BL, Mice, Knockout, Phosphoprotein Phosphatases
Abstract

Inhibition of AMPK is tightly associated with metabolic perturbations upon over nutrition, yet the molecular mechanisms underlying are not clear. Here, we demonstrate the serine/threonine-protein phosphatase 6 regulatory subunit 3, SAPS3, is a negative regulator of AMPK. SAPS3 is induced under high fat diet (HFD) and recruits the PP6 catalytic subunit to deactivate phosphorylated-AMPK, thereby inhibiting AMPK-controlled metabolic pathways. Either whole-body or liver-specific deletion of SAPS3 protects male mice against HFD-induced detrimental consequences and reverses HFD-induced metabolic and transcriptional alterations while loss of SAPS3 has no effects on mice under balanced diets. Furthermore, genetic inhibition of AMPK is sufficient to block the protective phenotype in SAPS3 knockout mice under HFD. Together, our results reveal that SAPS3 is a negative regulator of AMPK and suppression of SAPS3 functions as a guardian when metabolism is perturbed and represents a potential therapeutic strategy to treat metabolic syndromes.

Published
2023

33. Self-Support Few-Shot Semantic Segmentation

Author

Fan, Qi, Pei, Wenjie, Tai, Yu-Wing, and Tang, Chi-Keung

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Existing few-shot segmentation methods have achieved great progress based on the support-query matching framework. But they still heavily suffer from the limited coverage of intra-class variations from the few-shot supports provided. Motivated by the simple Gestalt principle that pixels belonging to the same object are more similar than those to different objects of same class, we propose a novel self-support matching strategy to alleviate this problem, which uses query prototypes to match query features, where the query prototypes are collected from high-confidence query predictions. This strategy can effectively capture the consistent underlying characteristics of the query objects, and thus fittingly match query features. We also propose an adaptive self-support background prototype generation module and self-support loss to further facilitate the self-support matching procedure. Our self-support network substantially improves the prototype quality, benefits more improvement from stronger backbones and more supports, and achieves SOTA on multiple datasets. Codes are at \url{https://github.com/fanq15/SSP}., Comment: ECCV 2022

Published
2022

34. Fine-Grained Modeling and Optimization for Intelligent Resource Management in Big Data Processing

Author

Lyu, Chenghao, Fan, Qi, Song, Fei, Sinha, Arnab, Diao, Yanlei, Chen, Wei, Ma, Li, Feng, Yihui, Li, Yaliang, Zeng, Kai, and Zhou, Jingren

Subjects
Computer Science - Databases, Computer Science - Distributed, Parallel, and Cluster Computing
Abstract

Big data processing at the production scale presents a highly complex environment for resource optimization (RO), a problem crucial for meeting performance goals and budgetary constraints of analytical users. The RO problem is challenging because it involves a set of decisions (the partition count, placement of parallel instances on machines, and resource allocation to each instance), requires multi-objective optimization (MOO), and is compounded by the scale and complexity of big data systems while having to meet stringent time constraints for scheduling. This paper presents a MaxCompute-based integrated system to support multi-objective resource optimization via fine-grained instance-level modeling and optimization. We propose a new architecture that breaks RO into a series of simpler problems, new fine-grained predictive models, and novel optimization methods that exploit these models to make effective instance-level recommendations in a hierarchical MOO framework. Evaluation using production workloads shows that our new RO system could reduce 37-72% latency and 43-78% cost at the same time, compared to the current optimizer and scheduler, while running in 0.02-0.23s.

Published
2022

37. Experiment Research on 600 Gbit/s Communication System

Author

CAO Liang, FAN Qi, and LIU Ting

Subjects
high-speed communication, incident power, EON, spectral efficiency, Applied optics. Photonics, TA1501-1820
Abstract

【Objective】The rapid development of data center networks necessitates intensive research into high-speed and high-capacity data center optical transmission systems. Single carrier 400 or 600 Gbit/s transmission will gradually become the mainstream transmission rate in the next generation of communication network. While the application of 400 Gbit/s transmission is well-studied in data center networks, detailed research on 600 Gbit/s transmission systems is less common.【Methods】This study is based on a single carrier 600 Gbit/s next-generation data center Elastic Optical Network (EON) transmission system, and conducts detailed theoretical analysis and experimental research on factors affecting transmission distance and spectrum utilization efficiency.【Results】The analysis shows that the maximum transmission distance is determined by the single carrier's peak input optical power, while the spectrum utilization efficiency is related to the bandwidth of the transmission channel. Experimental investigation is conducted on a single carrier 600 Gbit/s EON transmission system. By comparing the relationships between different fiber input optical powers, the system’s Q factor and the pre correction bit error rate, as well as the relationships between 3 dB channel filtering bandwidth and system’s Q factor for different channel numbers, it is confirmed that the system’s transmission distance and spectrum utilization efficiency are related to the optimal fiber input optical power and filtering bandwidth, respectively. Furthermore, the experiment shows that the optimal single-wave input optical power and optimal filtering bandwidth of the 600 Gbit/s transmission system are + 4 dBm and 77 GHz, respectively. Under these settings, the system achieves the longest transmission distance and the highest spectral efficiency. Under these experimental conditions, the 600 Gbit/s transmission system achieves long-term stable operation without error for 48 hours, indicating that this fiber input power and bandwidth can effectively extend and improve the transmission distance and spectrum utilization of the 600 Gbit/s communication system.【Conclusion】The 600 Gbit/s EON system has an optimal input power and filter bandwidth that maximizes bandwidth utilization and transmission distance without significant fiber nonlinear effects or channel interference. The findings have significant values for the engineering construction of 600 Gbit/s transmission systems.

Published
2024
Full Text
View/download PDF

38. Physiological Response of Haloxylon ammodendron Seedlings to Drought Stress Under Different Levels of Nitrogen Application

Author

Yaxin SHI, Ze WANG, Cai REN, Hanhan ZHENG, Cong ZHOU, and Fan QI

Subjects
haloxylon ammodendron seedlings, drought, nitrogen application, biomass, physiology, Agriculture
Abstract

【Objective】It aims to explore the response of different nitrogen fertilizer levels to the growth and physiological changes of Haloxylon ammodendron seedlings under drought stress, and to provide theoretical basis and technical parameters for ecological restoration and vegetation conservation management in desert areas.【Method】The H. ammodendron seedlings in the desert area of Xinjiang were taken as research objects, and the changes of natural drought (water) conditions of the soil were simulated in the critical periods of seedling growth through pot test in high temperature season, and treatments of CK, mild drought, moderate drought and severe drought were set up according to the saturated water holding capacity of the soil, and the nitrogen levels of N0, N1, N2, N3, N4 and N5 (0、0.05、2.50、5.00、7.50、10.00g/m2) were set up respectively under the treatment of each drought gradient. Field collection of each treatment was carried out. The whole plant was labelled, the assimilated branches of labelled seedlings were collected randomly, the biomass of seedlings was determined by indoor drying, and the contents of soluble protein, soluble sugar, proline and malondialdehyde in assimilated branches were analyzed and determined to study the physiological response characteristics of seedlings to drought changes under different nitrogen application levels.【Result】Under different drought gradients, taking mild drought as an example, the plant biomass of seedlings showed a trend of increasing and then decreasing with the increase of nitrogen application, and the plant biomass increased from 23.28 g at N0 to 30.19 g at N3 and then decreased to 27.57 g at N5, indicating that excessive application of nitrogen was not conducive to the accumulation of seedling biomass. The contents of soluble protein, soluble sugar, proline and malondialdehyde in the assimilated branches of seedlings increased with the increase of drought gradient at different N application levels. When the assimilated branches were under severe drought, the contents of soluble protein, soluble sugar, proline and malondialdehyde accumulated rapidly, they adapted to the drought through osmoregulation, and the contents of each of them in the assimilated branches at the N3 application level showed the smallest increase compared with that in CK, with the increase of 58.96%, 73.01%, 73.1% and 73.9%, respectively, indicating that reasonable nitrogen application could effectively alleviate drought.【Conclusion】Reasonable fertilization is beneficial to the accumulation of seedling plant biomass and effectively alleviate drought stress. It is found that, under the N3 nitrogen application level, the biomass of seedling plant under each drought treatment reached the maximum value, and the contents of soluble protein, soluble sugar, proline and malondialdehyde in the assimilated branches show the least increase with the increase of drought gradient. Considering the comprehensive effect of fertilizer application, it is recommended that 5.00 g/m2 of N fertilizer can be added to the conventional planting when artificially planting H. ammodendron transplants in the desert areas of Xinjiang.

Published
2024
Full Text
View/download PDF

39. Research progress of clinical application and complication of contact lens in keratoconus

Author

Ren Shengwei, Fan Qi, Yang Kaili, and Lyu Tianbin

Subjects
corneal contact lens, scleral contact lens, keratoconus, clinical application, complication, Ophthalmology, RE1-994
Abstract

Keratoconus is a progressive blinding eye disease that characterized by corneal thinning and protrusion, which accompanied with irregular astigmatism and impaired visual acuity.The irregular astigmatism of early keratoconus can be corrected by spectacles. For the irregular astigmatism of moderate to severe keratoconus, spectacles are no longer suitable, and contact lenses are the best choice for patients to restore vision. There are various types of contact lenses, making the selection very difficult. In addition, trying on lenses for a long time will increase the discomfort and overall feeling of patients, and greatly increase the workload of doctors. Thus, the article aims to summarize and discuss the classification of contact lenses, the application of contact lenses in different types of keratoconus, the complications of contact lens, and the current status and prospect of contact lenses, with a view to understanding the management and clinical application of contact lenses in keratoconus patients and to further improving the application value of contact lenses in keratoconus.

Published
2024
Full Text
View/download PDF

40. GCoNet+: A Stronger Group Collaborative Co-Salient Object Detector

Author

Zheng, Peng, Fu, Huazhu, Fan, Deng-Ping, Fan, Qi, Qin, Jie, Tai, Yu-Wing, Tang, Chi-Keung, and Van Gool, Luc

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

In this paper, we present a novel end-to-end group collaborative learning network, termed GCoNet+, which can effectively and efficiently (250 fps) identify co-salient objects in natural scenes. The proposed GCoNet+ achieves the new state-of-the-art performance for co-salient object detection (CoSOD) through mining consensus representations based on the following two essential criteria: 1) intra-group compactness to better formulate the consistency among co-salient objects by capturing their inherent shared attributes using our novel group affinity module (GAM); 2) inter-group separability to effectively suppress the influence of noisy objects on the output by introducing our new group collaborating module (GCM) conditioning on the inconsistent consensus. To further improve the accuracy, we design a series of simple yet effective components as follows: i) a recurrent auxiliary classification module (RACM) promoting model learning at the semantic level; ii) a confidence enhancement module (CEM) assisting the model in improving the quality of the final predictions; and iii) a group-based symmetric triplet (GST) loss guiding the model to learn more discriminative features. Extensive experiments on three challenging benchmarks, i.e., CoCA, CoSOD3k, and CoSal2015, demonstrate that our GCoNet+ outperforms the existing 12 cutting-edge models. Code has been released at https://github.com/ZhengPeng7/GCoNet_plus., Comment: T-PAMI 2023

Published
2022
Full Text
View/download PDF

42. Giant pulmonary artery aneurysm associated with SARS-CoV-2 infection and Actinomyces odontolyticus sepsis: A case report

Author

Ling Wang, Wenyi Lei, Fan Qi, and Zheyuan Fan

Subjects
Pulmonary artery aneurysm, COVID-19, SARS-CoV-2, Actinomyces odontolyticus, Case report, Science (General), Q1-390, Social sciences (General), H1-99
Abstract

Giant septic pulmonary artery aneurysms (PAAs) are rare but important entities, with few cases having been reported worldwide. Early diagnosis and prompt treatment are crucial in the management of such cases. We report a 56-year-old female patient presenting with fatigue, nausea and vomiting who was first diagnosed with diabetic ketoacidosis (DKA) and developed life-threatening giant infectious PAA secondary to SARS-CoV-2 infection and Actinomyces odontolyticus sepsis. The patient did not develop any specific symptoms, and enhanced computed tomography (CT) revealed a massive PAA of 5.6 × 4.9 cm in size at the left pulmonary hilar with normal pulmonary artery (PA) pressures. After multidisciplinary discussion and after considering the critical condition accompanied by sepsis increased the risk of surgery, endovascular treatment was the first therapy of choice for the patient; nevertheless, the patient ultimately opted for hospice care. This case report aims to raise awareness of PAAs, which are rare but potentially fatal complications of infectious diseases such as COVID-19 pneumonia and Actinomyces odontolyticus sepsis.

Published
2024
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results