A5101797846- Advanced Neural Network Applications
- Autonomous Vehicle Technology and Safety
- Molecular Biology Techniques and Applications
- Advanced Graph Neural Networks
- DNA and Biological Computing
- RNA Research and Splicing
- Genetic and phenotypic traits in livestock
- Cooperative Communication and Network Coding
Jilin University
2024
Background With the gradual rise of antibiotic-free farming practices, exploration novel, green, and low-pollution alternatives to antibiotics has become one key research focus in field agricultural science. In development antibiotic alternatives, probiotics, particularly host-associated have been found play a significant role enhancing production performance livestock poultry. However, on application probiotics specifically for meat pigeons remain relatively underdeveloped. Objective To...
Using generative models to synthesize new data has become a de-facto standard in autonomous driving address the scarcity issue. Though existing approaches are able boost perception models, we discover that these fail improve performance of planning end-to-end as generated videos usually less than 8 frames and spatial temporal inconsistencies not negligible. To this end, propose Delphi, novel diffusion-based long video generation method with shared noise modeling mechanism across multi-views...
Goose creates important economic value depending on their enrich nutrients of meat. Our previous study investigates potential candidate genes associated with variations in meat quality between Xianghai Flying (XHF) and Zi through genomic transcriptome integrated analysis. Screening five differential expression related to muscle development identified by the FST, XP-EHH RNA-seq breast from various geese. Among them, C1QTNF1 (C1q TNF protein 1), a gene unknown function goose, which observed...
Graph node classification with few labeled nodes presents significant challenges due to limited supervision. Conventional methods often exploit the graph in a transductive learning manner. They fail effectively utilize abundant unlabeled data and structural information inherent graphs. To address these issues, we introduce Structure-Aware Consensus Network (SACN) from three perspectives. Firstly, SACN leverages novel structure-aware consensus strategy between two strongly augmented views....