Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks40 from public data
- MeshMAE: Masked Autoencoders for 3D Mesh Data Analysis77
Inspired by MAE, this learning paradigm for 3D mesh data analysis based on Transformers is explored, which can yield state-of-the-art or comparable performance on mesh analysis tasks, i.e., classification and segmentation.
- GFNet: Geometric Flow Network for 3D Point Cloud Semantic Segmentation54
This paper introduces a geometric flow network (GFNet) to explore the geometric correspondence between different views in an align-before-fuse manner and devise a novel geometric flow module (GFM) to bidirectionally align and propagate the complementary information across different views according to geometric relationships under the end-to-end learning scheme.
- Toward Efficient Language Model Pretraining and Downstream Adaptation via Self-Evolution: A Case Study on SuperGLUE46
This technical report briefly describes the JDExplore d-team's Vega v2 submission on the SuperGLUE leaderboard, which achieved new state-of-the-art performance on 4/8 tasks, and the prompt transfer technique to improve the low-resource tasks by transferring the knowledge from the foundation model and related downstream tasks to the target task.
- Knowledge-Aware Federated Active Learning with Non-IID Data36
This paper proposes Knowledge-Aware Federated Active Learning (KAFAL), which consists of Knowledge-Specialized Active Sampling (KSAS) and Knowledge-Compensatory Federated Update (KCFU), a novel active sampling method tailored for the federated active learning problem.
- Improving Fine-Grained Visual Recognition in Low Data Regimes via Self-Boosting Attention Mechanism35
The self-boosting attention mechanism is proposed, a novel method for regularizing the network to focus on the key regions shared across samples and classes to significantly improve fine-grained visual recognition performance on low data regimes and can be incorporated into existing network architectures.
- Discovering Human-Object Interaction Concepts via Self-Compositional Learning27
A novel and challenging task for a comprehensive HOI understanding, which is termed as HOI Concept Discovery, and a self-compositional learning framework (or SCL) for HOI concept discovery is introduced, which enables the learning on both known and unknown HOI concepts.
- Neighborhood relation-based knowledge distillation for image classification22
This paper first finds a subset of samples with their K-nearest neighbors according to the similarity matrix of mini-batch samples and then builds the neighborhood relationship knowledge for knowledge distillation, where the characterized relational knowledge can be transferred by both intermediate feature maps and output logits.
- On Exploring Node-feature and Graph-structure Diversities for Node Drop Graph Pooling19
A groundbreaking plug-and-play score scheme, termed MID, which comprises a Multidimensional score space and two key operations: flIpscore and Dropscore, which has proven to bring a significant average improvement over existing node drop pooling methods when tested on 17 real-world graph classification datasets.
- Pseudo Contrastive Learning for Graph-based Semi-supervised Learning16
A general framework for GNNs, termed Pseudo Contrastive Learning (PCL), which separates two nodes whose positive and negative pseudo-labels target the same class, and devise a topologically weighted contrastive loss that spends more effort separating negative pairs with smaller topological distances.
- BatchFormerV2: Exploring Sample Relationships for Dense Representation Learning16
A more general batch Transformer module, BatchFormerV2, which further enables exploring sample relationships for dense representation learning and consistently improves current DETR-based detection methods by over 1.3%.
- 15
- 13
- Compositional 3D Human-Object Neural Animation12
A new compositional conditional neural radiance field (or CC-NeRF) is devised, which decomposes the interdependence between human and object using latent codes to enable compositionally animation control of novel HOIs.
- 11
- 11
- Progressive Retinal Image Registration via Global and Local Deformable Transformations10
A hybrid registration framework called HybridRetina is proposed, which progressively registers retinal images with global and local deformable transformations, and uses a keypoint detector and a deformation network called GAMorph to estimate the global and local deformable transformation, respectively.
- ChartDETR: A Multi-shape Detection Network for Visual Chart Recognition10
This work proposes ChartDETR, a transformer-based multi-shape detector that localizes keypoints at the corners of regular shapes to reconstruct multiple data elements in a single chart image, and predicts all data element shapes at once by introducing query groups in set prediction, eliminating the need for further postprocessing.
- Sini decoction intervention on atherosclerosis via PPARγ-LXRα-ABCA1 pathway in rabbits10
SND treatment relieved AS, improved lipid profiles, and increased serum HDL-C level, which might be the improvement of reverse cholesterol transport (RCT) involved with enhanced expression of ABCA1, ApoA-I, PPARγ, and LXRα.
- Collect-and-Distribute Transformer for 3D Point Cloud Analysis8
A new transformer network equipped with a collect-and-distribute mechanism to communicate short- and long-range contexts of point clouds, which is referred to as CDFormer is proposed, delivering several new state-of-the-art performances on point cloud classification and segmentation tasks.
- 6
- 4
- 4
- Research and verification of sensing information assisted millimeter wave beam tracking algorithm for automated vehicles2
A sensing information assisted mmWave beam tracking algorithm combining both camera and light detection and ranging (LiDAR) is proposed, which can guarantee the high data rate communication performance regardless of sufficient and insufficient ambient light under the mobility scenarios.
- Instruction Learning Paradigms: A Dual Perspective on White-Box and Black-Box LLMs1
This framework aligns black-box and white-box representations within a shared semantic space and incorporates a feature adaptation module to dynamically capture inter-model consistency and complementarity, thereby enhancing information fusion and semantic robustness.
- Re-Initialization Token Learning for Tool-Augmented Large Language Models1
A novel token learning method that aligns tool tokens with the existing word embedding space from the perspective of initialization, thereby enhancing model performance and effectively augments LLMs with tools through relevant tokens across diverse domains.
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; 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.