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
- Nenghai YuSuggested from co-authorship
- Baining GuoSuggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works49 from public data
- CSWin Transformer: A General Vision Transformer Backbone with Cross-Shaped Windows1,458
The Cross-Shaped Window self-attention mechanism for computing self-Attention in the horizontal and vertical stripes in parallel that form a cross-shaped window is developed, with each stripe obtained by splitting the input feature into stripes of equal width.
- Mobile-Former: Bridging MobileNet and Transformer727
The proposed light-weight cross attention to model the bridge enables bidirectional fusion of local and global features and builds an efficient end-to-end detector by replacing backbone, encoder and decoder in DETR with Mobile-Former, which outperforms DETR but saves 52% of computational cost and 36% of parameters.
- VLMEvalKit: An Open-Source ToolKit for Evaluating Large Multi-Modality Models628
The VLMEvalKit toolkit aims to provide a user-friendly and comprehensive framework for researchers and developers to evaluate existing multi-modality models and publish reproducible evaluation results.
- Open-Sora Plan: Open-Source Large Video Generation Model325
This work introduces Open-Sora Plan, an open-source project that aims to contribute a large generation model for generating desired high-resolution videos with long durations based on various user inputs, and hopes it can inspire the video generation research community.
- PeCo: Perceptual Codebook for BERT Pre-training of Vision Transformers289
A self-supervised transformer model for deep feature extraction and it is shown that it works well for calculating perceptual similarity, and such learned visual tokens indeed exhibit better semantic meanings, and help pre-training achieve superior transfer performance in various downstream tasks.
- 237
- InternLM-XComposer-2.5: A Versatile Large Vision Language Model Supporting Long-Contextual Input and Output216
A versatile large-vision language model that supports long-contextual input and output, InternLM-XComposer-2.5 excels in various text-image comprehension and composition applications, achieving GPT-4V level capabilities with merely 7B LLM backend.
- OVO-Bench: How Far is Your Video-LLMs from Real-World Online Video Understanding?159
OVO-Bench evaluates the ability of video LLMs to reason and respond to events occurring at specific timestamps under three distinct scenarios: Backward tracing, real-time understanding, and forward active responding.
- LG-GAN: Label Guided Adversarial Network for Flexible Targeted Attack of Point Cloud Based Deep Networks130
The proposed LG-GAN can support flexible targeted attack on the fly while guaranteeing good attack performance and higher efficiency simultaneously, and is the first generation based 3D point cloud attack method.
- Shape-invariant 3D Adversarial Point Clouds120
A novel Point-Cloud Sensitivity Map is proposed to boost both the efficiency and imperceptibility of point perturbations and reveals the vulnerability of point cloud recognition models when encountering shape-invariant adversarial noises.
- 111
- 103
- Local Geometric Distortions Resilient Watermarking Scheme Based on Symmetry85
Experimental results demonstrate the excellent robustness of the proposed watermarking scheme to local geometric distortion, global geometric distortions, common image processing operations, and some kinds of combined attacks.
- 83
- 72
- 65
- 64
- Spatial-SSRL: Enhancing Spatial Understanding via Self-Supervised Reinforcement Learning44
Spatial-SSRL is introduced, a self-supervised RL paradigm that derives verifiable signals directly from ordinary RGB or RGB-D images and provides a practical route to stronger spatial intelligence in LVLMs.
- 43
- Maximum Entropy Reinforcement Learning with Diffusion Policy41
The diffusion model is employed, a powerful generative model capable of capturing complex multimodal distributions, as the policy representation to fulfill the MaxEnt RL objective, developing a method named MaxEnt RL with Diffusion Policy (MaxEntDP), which enables efficient exploration and brings the policy closer to the optimal MaxEnt policy.
- Beyond Fixed: Training-Free Variable-Length Denoising for Diffusion Large Language Models36
DAEDAL is introduced, a novel training-free denoising strategy that enables Dynamic Adaptive Length Expansion for Diffusion Large Language Models and achieves performance comparable, and in some cases superior, to meticulously tuned fixed-length baselines, while simultaneously enhancing computational efficiency by achieving a higher effective token ratio.
- Once a MAN: Towards Multi-Target Attack via Learning Multi-Target Adversarial Network Once33
This paper proposes the first Multi-target Adversarial Network (MAN), which can generate multi-target adversarial samples with a single model and can achieve better classification accuracy than other methods when attacked by various methods.
- RAR: Retrieving and Ranking Augmented MLLMs for Visual Recognition31
RAR, a Retrieving And Ranking augmented method for Multimodal Large Language Models, demonstrates a significant improvement in performance on 5 fine-grained visual recognition benchmarks, 11 few-shot image recognition datasets, and the 2 object detection datasets under the zero-shot recognition setting.
- 30
- 30
- TACR-Net: Editing on Deep Video and Voice Portraits27
A novel deep learning framework, named Temporal-Refinement Autoregressive-Cascade Rendering Network (TACR-Net) for audio-driven dynamic talking face editing, which encodes facial expression blendshape based on the given acoustic features without separately training for special video.
- Adversarial steganography based on sparse cover enhancement27
A novel cover enhancement method, denoted as SPS-ENH (SParSe ENHancement), which effectively compress the distortions caused in cover enhancement, and outperforms the previous works in the average classification error rates under non-target steganalyzers and deceiving success rates against target CNN models.
- 26
- BoostStep: Boosting mathematical capability of Large Language Models via improved single-step reasoning22
BoostStep is a method that enhances reasoning accuracy through step-aligned ICL, a novel mechanism that carefully aligns retrieved reference steps with the corresponding reasoning steps, and enhances state-of-the-art LLMs to solve challenging math problems using simpler examples.
- Emotional Listener Portrait: Realistic Listener Motion Simulation in Conversation22
The Emotional Listener Portrait (ELP) model can not only automatically generate natural and diverse responses toward a given speaker via sampling from the learned distribution but also generate controllable responses with a predetermined attitude.
- 22
- PersonMAE: Person Re-Identification Pre-Training With Masked AutoEncoders18
A simple yet effective pre-training framework, namely PersonMAE, which involves two core designs into masked autoencoders to better serve the task of Person Re-ID, leading to State-of-the-Art performance on four downstream ReID tasks, i.e., supervised (holistic and occluded setting), and unsupervised (UDA and USL setting).
- Improving Adversarial Robustness of Masked Autoencoders via Test-time Frequency-domain Prompting18
This paper provides a simple yet effective way to boost MAE’s adversarial robustness by using the dataset-extracted domain knowledge to occupy the medium-/high-frequency of images, thus narrowing the optimization space of adversarial perturbations.
- 16
- 13
- 12
- Demo-ICL: In-Context Learning for Procedural Video Knowledge Acquisition10
Demo-ICL, an MLLM with a two-stage training strategy: video-supervised fine-tuning and information-assisted direct preference optimization, jointly enhancing the model's ability to learn from in-context examples is developed.
- SPARK: Synergistic Policy And Reward Co-Evolving Framework10
The Synergistic Policy And Reward Co-Evolving Framework (SPARK), an efficient, on-policy, and stable method that builds on RLVR, is introduced, an efficient, on-policy, and stable method that eliminates the need for a separate reward model and costly human preference data.
- Adversarial defense via self-orthogonal randomization super-network10
This paper proposes a novel type of defense method named ‘‘Self-Orthogonal Randomization Super-network” ( SORS), which can achieve significantly better robustness than the vanilla single model baseline and the naive model ensemble baseline.
- CapRL++: Unified Reinforcement Learning with Verifiable Rewards for Dense Image and Video Captioning8
Results validate that CapRL++ effectively trains models to produce generalizable, high-fidelity descriptions, establishing a robust foundation beyond the limitations of traditional SFT.
- ByTheWay: Boost Your Text-to-Video Generation Model to Higher Quality in a Training-free Way7
ByTheWay is a training-free method to improve the quality of text-to-video generation without introducing additional parameters, augmenting memory or sampling time, and demonstrates that ByTheWay significantly improves the quality of text-to-video generation with negligible additional cost.
- 7
- Visual Self-Refine: A Pixel-Guided Paradigm for Accurate Chart Parsing6
This work proposes a new paradigm named Visual Self-Refine (VSR), to enable a model to generate pixel-level localization outputs, visualize them, and then feed these visualizations back to itself, allowing it to intuitively inspect and correct its own potential visual perception errors.
- 5
- 4
- VISTA: Enhancing Visual Conditioning via Track-Following Preference Optimization in Vision-Language-Action Models3
This work study VLA models through the lens of visual conditioning and empirically show that successful rollouts consistently exhibit stronger visual dependence than failed ones, and proposes a training framework that explicitly strengthens visual conditioning in VLA models.
- 1
- 1
- Detecting AGNs using multi-filter imaging data–
An imaging manipulating algorithm was applied to each galaxy image which allowed the reconstructed images to have the same PSF while enhancing the spatial resolution, and the results indicated that in the local Universe AGNs and passive galaxies prefer early-type galaxies, while SBs/SFs were more often found in late-type spirals.
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 2026-10-15. 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.