Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileAcademic lineage
Possible advisorsa guess from early papers, not confirmed
- Jun LiuSuggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works20 from public data
- LTGC: Long-Tail Recognition via Leveraging LLMs-Driven Generated Content55
A novel generative and fine-tuning framework to handle long-tail recognition via leveraging generated content inspired by the rich implicit knowledge in large-scale models, LTGC, which outperforms existing state-of-the-art methods on popular long-tailed benchmarks.
- MDCS: More Diverse Experts with Consistency Self-distillation for Long-tailed Recognition45
The method can effectively increase the diversity of experts, significantly reduce the variance of the model, and improve recognition accuracy, and experiments show the MDCS outperforms the state-of-the-art on five popular long-tailed benchmarks.
- MixPro: Data Augmentation with MaskMix and Progressive Attention Labeling for Vision Transformer23
MaskMix and Progressive Attention Labeling in image and label space are proposed and combined as a new data augmentation method, named MixPro, which can improve various ViT-based models at scales on ImageNet classification (73.8\% top-1 accuracy based on DeiT-T for 300 epochs).
- PEACE: Empowering Geologic Map Holistic Understanding with MLLMs16
GeoMap-Agent, the inaugural agent designed for geologic map understanding, features three modules: Hierarchical Information Extraction (HIE), Domain Knowledge Injection (DKI), and Prompt-enhanced Question Answering (PEQA), which paves the way for advanced AI applications in geology, enhancing the efficiency and accuracy of geological investigations.
- GaussianBlock: Building Part-Aware Compositional and Editable 3D Scene by Primitives and Gaussians12
This paper proposes a novel part-aware compositional reconstruction method, called GaussianBlock, that enables semantically coherent and disentangled representations, allowing for precise and physical editing akin to building blocks, while simultaneously maintaining high fidelity.
- P-DIFF+: Improving learning classifier with noisy labels by Noisy Negative Learning loss12
This paper presents a very simple but effective training paradigm called P-DIFF+, which can train DNN classifiers but obviously alleviate the adverse impact of noisy labels, and its proposed probability difference distribution implicitly reflects the probability of a training sample to be clean, then this probability is employed to re-weight the corresponding sample during the training process.
- 10
- 9
- P-DIFF: Learning Classifier with Noisy Labels based on Probability Difference Distributions9
A very simple but effective training paradigm called P-DIFF, which can train DNN classifiers but obviously alleviate the adverse impact of noisy labels, and is superior to the state-of-the-art sample selection methods.
- 8
- 3
- ResearchStudio-Reel: Automate the Last Mile of Research from Paper to Poster, Video, and Blog2
ResearchStudio-Reel is presented, a native-editable dissemination workspace that binds its three artifacts into one interactive deliverable at the experience level, implemented as five skills executable in Claude Code and Codex: one shared extractor, three editable artifact generators, and one interactive convergence layer.
- Polybrominated diphenyl ether profiles in adipose tissues of breast cancer patients and their carcinogenic potential investigation based on network toxicology and molecular docking2
This integrative network study uncovers a mechanistic framwork linking adipose-accumulated PBDE mixtures to breast cancer pathogenesis and provides insights for preventive and therapeutic interventions against PBDE-associated breast cancer.
- 2
- Editorial: Enhancing T cell function: innovations in cancer immunotherapy1
Functional assays revealed that three clinically relevant T-cell subpopulations predicted reduced risks of secondary malignancies and mortality, validated across cohorts.
- –
- –
- Prediction of battery performance degradation based on machine learning–
The ideas put forward in this paper can be used for predictive maintenance and abnormal replacement of fuel cells, and also have certain reference significance for similar lithium-ion batteries.
- –
- –
Publication data from OpenAlex; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-11. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it. You will be asked to sign in.