Lu Jiang

ORCID: 0000-0001-8930-0488
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Retinoids in leukemia and cellular processes
  • Acute Myeloid Leukemia Research
  • Advanced biosensing and bioanalysis techniques
  • Cancer, Lipids, and Metabolism
  • Generative Adversarial Networks and Image Synthesis
  • Probiotics and Fermented Foods
  • Angiogenesis and VEGF in Cancer
  • Peroxisome Proliferator-Activated Receptors
  • Cancer Research and Treatments
  • Oral Health Pathology and Treatment
  • Turtle Biology and Conservation
  • Spider Taxonomy and Behavior Studies
  • Biopolymer Synthesis and Applications
  • Protein Degradation and Inhibitors
  • Subterranean biodiversity and taxonomy
  • Multimodal Machine Learning Applications
  • Head and Neck Cancer Studies
  • Urban Heat Island Mitigation
  • Phytochemicals and Antioxidant Activities
  • Advanced Image and Video Retrieval Techniques
  • Molecular Biology Techniques and Applications
  • Endometriosis Research and Treatment
  • Enzyme Catalysis and Immobilization
  • Oral and gingival health research
  • Gut microbiota and health

Suzhou Municipal Hospital
2025

Nanjing Medical University
2025

Beijing Normal University
2010-2025

Chengdu University
2023-2024

Anhui University of Traditional Chinese Medicine
2022-2024

Tianjin Agricultural University
2022-2024

Shandong First Medical University
2022-2024

Guizhou University
2022-2024

Guizhou Academy of Agricultural Sciences
2022-2024

First Affiliated Hospital of Fujian Medical University
2024

We present Muse, a text-to-image Transformer model that achieves state-of-the-art image generation performance while being significantly more efficient than diffusion or autoregressive models. Muse is trained on masked modeling task in discrete token space: given the text embedding extracted from pre-trained large language (LLM), to predict randomly tokens. Compared pixel-space models, such as Imagen and DALL-E 2, due use of tokens requiring fewer sampling iterations; compared Parti,...

10.48550/arxiv.2301.00704 preprint EN cc-by arXiv (Cornell University) 2023-01-01

Accurate medical image segmentation is essential for clinical quantification, disease diagnosis, treatment planning and many other applications. Both convolution-based transformer-based u-shaped architectures have made significant success in various tasks. The former can efficiently learn local information of images while requiring much more image-specific inductive biases inherent to convolution operation. latter effectively capture long-range dependency at different feature scales using...

10.48550/arxiv.2401.00722 preprint EN other-oa arXiv (Cornell University) 2024-01-01

Pre-trained large text-to-image models synthesize impressive images with an appropriate use of text prompts. However, ambiguities inherent in natural language and out-of-distribution effects make it hard to image styles, that leverage a specific design pattern, texture or material. In this paper, we introduce StyleDrop, method enables the synthesis faithfully follow style using model. The proposed is extremely versatile captures nuances details user-provided style, such as color schemes,...

10.48550/arxiv.2306.00983 preprint EN cc-by arXiv (Cornell University) 2023-01-01

10.1016/j.cytogfr.2024.03.001 article EN Cytokine & Growth Factor Reviews 2024-03-04

High endothelial venules (HEV) are specialized post-capillary that recruit naïve lymphocytes to lymph nodes. HEVs essential for the development of adaptive immunity. can also develop in tumors where they thought be important recruiting T cells and B into locally enhancing antitumor immunity by supporting formation tertiary lymphoid structures. Herein, we used comparative transcriptome analysis human breast cancer investigate genes differentially expressed between tumor-associated rest tumor...

10.1158/2326-6066.cir-21-0369 article EN cc-by-nc-nd Cancer Immunology Research 2022-02-21

ABSTRACT The Qinghai‐Tibet Plateau (QTP) has an extensive frozen soil distribution and intense geological tectonic activity. Our surveys reveal that earthquakes can not only damage infrastructure but also significantly impact carbon dioxide emissions. Fissures created by expose deep, soils to the air and, in turn, accelerate We measured average emission rates of 968.53 g CO 2 m −2 ·a −1 on fissure sidewall 514.79 at bottom. estimated total flux from fissures caused M ≥ 6.9 326 B.C. 2022 is...

10.1111/gcb.70024 article EN cc-by Global Change Biology 2025-01-01

Abstract Blueberries ( Vaccinium spp.) are susceptible to mechanical damage and fungal diseases during storage. To explore the effect of strigolactone treatment on preservation blueberries, ‘Bluerain’ variety was treated with rac -GR24, a synthetic postharvest, various physiological biochemical indicators were analysed. The results showed that, compared control, GR24 significantly inhibited increase in weight-loss rate, decay incidence, malondialdehyde (MDA) content H 2 O blueberries...

10.1079/ejhs.2025.0002 article EN European Journal of Horticultural Science 2025-02-14

Deciphering human behaviors to predict their future paths/trajectories and what they would do from videos is important in many applications. Motivated by this idea, paper studies predicting a pedestrian's path jointly with activities. We propose an end-to-end, multi-task learning system utilizing rich visual features about behavioral information interaction surroundings. To facilitate the training, network learned auxiliary task of location which activity will happen. Experimental results...

10.48550/arxiv.1902.03748 preprint EN other-oa arXiv (Cornell University) 2019-01-01

The study on the bacterial community structure and metabolite effects of fermentation Yu jiangsuan(YJS)by autochthonous starter cultures has been rarely reported. In this study, metabolites were dynamically explored by high-throughput sequencing LC-MS/MS during jiangsuan (YJS) inoculation cultures. results demonstrated that presence interactions between genera accelerated process caused an increase in diversity production. Firmicutes, Cyanobacteria Proteobacteria dominant phyla, while...

10.1016/j.lwt.2022.113874 article EN cc-by-nc-nd LWT 2022-08-22

Conjugation of carbohydrates to ferrocene scaffolds is great value in drug design, given the nontoxic and lipophilic nature ferrocene. However, efficient stereoselective synthesis C-ferrocenyl glycosides remains a challenge. Herein, we developed Pd-catalyzed C-H glycosylation provide rapid access sole bis-C-ferrocenyl good high yields (up 98% yield) with exclusive stereoselectivity. A diverse range glycosyl chlorides were well tolerated, including d-mannose, d-glucose, l-xylose, l-rhamnose,...

10.1021/acs.orglett.3c01266 article EN Organic Letters 2023-05-26

Fatty acids are one of the main sources flavour in fermented Yu jiaosuan (YJS) southwest China. Bacilli (50.18 %) and Oxyphotobacteria (32.70 were dominant class. Lactiplantibacillus (40.51 Weissella (20.43 species inoculated group (HY). The peroxide value (ZY: 0.025 g/100 g, HY: 0.016 g) lipoxygenase (LOX) 5.7654 U/min·g, 3.3856 U/min·g) HY significantly lower compared with natural fermentation (ZY), while acid lipase activity 0.3184 U/h·g, 0.7075 U/h·g) neutral 12.65443 20.25142 higher...

10.1016/j.fochx.2024.101246 article EN cc-by-nc-nd Food Chemistry X 2024-02-19

Devising a low-cost and effective strategy to reduce Cd contamination of brown rice is critical achieve the safe production grain for human consumption. Accordingly, here field experiments were conducted at two sites evaluate effects applying humic acid (HA) foliage twice, booting full heading stage, on diminishing translocation cadmium (Cd) into grains. Besides measuring subcellular distribution in flag leaf polysaccharide composition cell wall, latter's types concentrations functional...

10.1016/j.ecoenv.2022.113658 article EN cc-by-nc-nd Ecotoxicology and Environmental Safety 2022-05-19

Abstract Background: GQ1001 was developed to overcome the current limitation of existing HER2 ADCs by using next generation site-specific conjugation and a stable linker technology, coined intelligent ligase-dependent (iLDC) technology. iLDC provides robust platform for ADC with high homogeneity excellent stability. Our data show that is highly homogeneous strong anti-tumor potency, more importantly, markedly enhanced stability reduced off-target toxicity, points safer larger therapeutic...

10.1158/1538-7445.am2023-2702 article EN Cancer Research 2023-04-04

We present VideoPoet, a language model capable of synthesizing high-quality video, with matching audio, from large variety conditioning signals. VideoPoet employs decoder-only transformer architecture that processes multimodal inputs -- including images, videos, text, and audio. The training protocol follows Large Language Models (LLMs), consisting two stages: pretraining task-specific adaptation. During pretraining, incorporates mixture generative objectives within an autoregressive...

10.48550/arxiv.2312.14125 preprint EN cc-by arXiv (Cornell University) 2023-01-01
Coming Soon ...