Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks25 from public data
- 87
- 54
- 35
- 34
- Adaptive Perception for Unified Visual Multimodal Object Tracking25
This work proposes APTrack, a novel unified tracker designed for multimodal adaptive perception that surpasses existing state-of-the-art unified multimodal trackers but also outperforms trackers designed for specific multimodal tasks.
- 22
- 15
- SwimVG: Step-Wise Multimodal Fusion and Adaption for Visual Grounding12
Swip and CIA are both parameter-efficient paradigms, and they fuse the cross-modal features from shallow to deep layers gradually, and they fuse the cross-modal features from shallow to deep layers gradually.
- Fast Adversarial Training With Weak-to-Strong Spatial-Temporal Consistency in the Frequency Domain on Videos10
Video Fast Adversarial Training with Weak-to-Strong consistency (VFAT-WS) is introduced, the first fast adversarial training method for video data that seamlessly integrates the simple TF-AUG and the more complex STF-AUG and steers the learning process from simple to complex augmentations.
- 9
- Explicit Context Reasoning with Supervision for Visual Tracking8
RSTrack is proposed, which explicitly models and supervises context reasoning via three core mechanisms, and achieves state-of-the-art performance on multiple benchmark datasets while maintaining real-time running speeds.
- 4
- An Effective End-to-End Solution for Multimodal Action Recognition4
This work proposes a comprehensive multimodal action recognition solution that effectively utilizes multimodal information and achieves the Top-1 accuracy of 99% and the Top-5 accuracy of 100% on the competition leaderboard, demonstrating the superiority of this solution.
- When Color-Space Decoupling Meets Diffusion for Adverse-Weather Image Restoration4
A novel framework comprising two key components: Lumina-Chroma Decomposition Network (LCDN) and Lumina-Guided Diffusion Model (LGDM), which surpasses state-of-the-art methods in AWIR and presents DriveWeather, a comprehensive all-weather driving dataset designed to enable robust evaluation.
- 3
- 3
- Personalized federated learning based on feature fusion2
This work replaces traditional gradient uploading with feature uploading, and introduces a novel feature fusion scheme to learn personalized local model for clients, which outperforms recent FL methods on the MNIST, FEMNIST, and CIFAR-10 datasets while requiring less communication.
- 1
- Probabilistic Memory for Trustworthy Edge Intelligence1
Probabilistic memory (p-MEM), a unified memory primitive that stores distribution parameters, such as mean and standard deviation, and samples directly at the native memory bandwidth, where deterministic data becomes the zero-variance special case is introduced.
- An Enhanced Multi-Target Detection Model for Autonomous Driving Based on Deep Learning Optimization–
An enhanced object detection solution with the following key innovations is proposed, replacing traditional convolutions in the backbone network with the T-CMUNeXt module to overcome computational inefficiency and shallow feature extraction, thereby providing high-quality initial feature maps.
- Artificial Intelligence in English Speaking Instruction: Learner Impacts and Teacher Role Reconstruction–
A dual influence model is revealed: AI effectively supports the development of basic oral skills and learning autonomy, but still has limitations in cultivating higher-level communicative and pragmatic abilities.
- –
- –
- –
- –
Publication data from OpenAlex; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-15. 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.