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
- Wanli OuyangSuggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works64 from public data
- Feature Erasing and Diffusion Network for Occluded Person Re-Identification207
A novel Feature Erasing and Diffusion Network (FED) to simultaneously handle challenges from NPO and NTP is proposed and the superiority of FED over state-of-the-art methods is demonstrated.
- 113
- Complementary Relation Contrastive Distillation109
This paper proposes a novel knowledge distillation method, namely Complementary Relation Contrastive Distillation (CRCD), to transfer the structural knowledge from the teacher to the student and estimates the mutual relation in an anchor-based way and distill the anchorstudent relation under the supervision of its corresponding anchor-teacher relation.
- Adapting Object Detectors with Conditional Domain Normalization103
The Conditional Domain Normalization (CDN) is designed to encode different domain inputs into a shared latent space, where the features from different domains carry the same domain attribute.
- Layerwise Optimization by Gradient Decomposition for Continual Learning100
This work decomposed the gradient of an old task into a part shared by all old tasks and a part specific to that task, and performs optimization for the gradients of each layer separately rather than the concatenation of all gradients as in previous works.
- UniHCP: A Unified Model for Human-Centric Perceptions95
This work proposes UniHCP, a Unified Model for Human-Centric Perceptions, which unifies a wide range of human-centric tasks in a simplified end-to-end manner with the plain vision transformer architecture and can outperform strong baselines on several in-domain and downstream tasks by direct evaluation.
- UniPAD: A Universal Pre-Training Paradigm for Autonomous Driving90
UniPAD is presented, a novel self-supervised learning paradigm applying 3D volumetric differentiable rendering that implicitly encodes 3D space, facilitating the reconstruction of continuous 3D shape structures and the intricate appear-ance characteristics of their 2D projections.
- Many Heads Are Better Than One: Improved Scientific Idea Generation by A LLM-Based Multi-Agent System88
This work proposes an LLM-based multi-agent system, i.e., Virtual Scientists (VirSci), designed to mimic the teamwork inherent in scientific research, and investigates the collaboration mechanisms that contribute to its tendency to produce ideas with higher novelty.
- Instruct-ReID: A Multi-Purpose Person Re-Identification Task with Instructions84
A new instruct-ReID task that requires the model to retrieve images according to the given image or language instructions, and a large-scale OmniReID benchmark and an adaptive triplet loss as a baseline method to facilitate research in this new setting.
- Fast crystal growth at ultra-low temperatures76
It is shown that the intrinsic mechanical instability of a disordered glassy state subject to the crystal growth front allows for domino-like fast crystal growth even at ultra-low temperatures.
- 75
- Gradient Regularized Contrastive Learning for Continual Domain Adaptation73
This work studies the problem of continual domain adaptation, where the model is presented with a labelled source domain and a sequence of unlabelled target domains, and proposes Gradient Regularized Contrastive Learning (GRCL), a method that constraining the gradient not to increase the classification loss on old target domains enables the model to preserve the performance onold target domains when adapting to an in-coming target domain.
- 69
- 66
- Revisiting the Transferability of Supervised Pretraining: an MLP Perspective62
This paper attempts to close the transferability gap between supervised and unsupervised pretraining by adding an MLP projector before the classifier in supervised pretraining, and indicates that theMLP projector can help retain intra-class variation of visual features, decrease the feature distribution distance between pretraining and evaluation datasets, and reduce feature redundancy.
- HumanBench: Towards General Human-Centric Perception with Projector Assisted Pretraining61
A HumanBench based on existing datasets is proposed to comprehensively evaluate on the common ground the generalization abilities of different pretraining methods on 19 datasets from 6 diverse downstream tasks, including person ReID, pose estimation, human parsing, pedestrian attribute recognition, pedestrian detection, and crowd counting.
- Continual Representation Learning for Biometric Identification55
This paper proposes a new continual learning setting, namely "continual representation learning", which focuses on learning better representation in a continuous way, and demonstrates that existing CL methods can improve the representation in the new setting.
- Interleaving Reasoning for Better Text-to-Image Generation40
This work introduces Interleaving Reasoning Generation (IRG), a framework that alternates between text-based thinking and image synthesis: the model first produces a text-based thinking to guide an initial image, then reflects on the result to refine fine-grained details, visual quality, and aesthetics while preserving semantics.
- A Survey of Scientific Large Language Models: From Data Foundations to Agent Frontiers31
A comprehensive, data-centric synthesis that reframes the development of Sci-LLMs as a co-evolution between models and their underlying data substrate, and outlines a paradigm shift toward closed-loop systems where autonomous agents based on Sci-LLMs actively experiment, validate, and contribute to a living, evolving knowledge base.
- 15
- Beyond Entropy: Region Confidence Proxy for Wild Test-Time Adaptation14
This paper introduces a novel region-integrated method ReCAP, a probabilistic region modeling scheme that flexibly captures semantic changes in embedding space and develops a finite-to-infinite asymptotic approximation that transforms the intractable region confidence into a tractable and upper-bounded proxy.
- 14
- Holistic-Motion2D: Scalable Whole-body Human Motion Generation in 2D Space13
This paper presents Holistic-Motion2D, the first comprehensive and large-scale benchmark for 2D whole-body motion generation, which includes over 1M in-the-wild motion sequences, each paired with high-quality whole-body/partial pose annotations and textual descriptions.
- 13
- 12
- 12
- Advances in 3D pre-training and downstream tasks: a survey9
This paper provides a comprehensive survey of recent advances in the 3D feld, which encompasses a wide collection of topics, including diverse pre-training strategies, backbone designs and downstream tasks.
- LEAD: Exploring Logit Space Evolution for Model Selection9
LEAD proposes a theoretical framework to model the optimization process and derives an ordinary differential equation (ODE) to depict the nonlinear evolution toward the final logit state, and designs a class-aware decomposition method to consider the varying evolution dynamics across classes and further ensure practical applicability.
- 7
- 6
- AI-Driven Automation Can Become the Foundation of Next-Era Science of Science Research5
A preliminary multi-agent system is presented as an illustrative example to simulate research societies, showcasing AI's ability to replicate real-world research patterns and accelerate progress in Science of Science research.
- 4
- 4
- 4
- 3
- 3
- 3
- 2
- 2
- 2
- 2
- 2
- 2
- 1
- 1
- 1
- Adaptive Dual Uncertainty Optimization: Boosting Monocular 3D Object Detection under Test-Time Shifts1
Dual Uncertainty Optimization (DUO) is proposed, the first TTA framework designed to jointly minimize both uncertainties for robust M3OD, and extensive experiments demonstrate the superiority of DUO over existing methods across various datasets and domain shift types.
- 1
- 1
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
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 2026-10-13. 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.