📝 Publications

CVPR 2026

Predictive Regularization Against Visual Representation Degradation in Multimodal Large Language Models
Enguang Wang, Qiang Wang, Yuanchen Wu, Ke Yan, Xinbin Yuan, Shouhong Ding, Xialei Liu, Ming-Ming Cheng

This paper conducts a systematic diagnosis to unveil the visual representation degradation phenomenon in MLLMs, identifying this phenomenon as a visual sacrifice driven by the singular text-generation objective.
It introduces Predictive Regularization (PRe), a lightweight self-supervised framework that preserves core visual competence by forcing degraded intermediate features to predict initial visual anchors.

CVPR 2025

GET: Unlocking the Multi-modal Potential of CLIP for Generalized Category Discovery
Enguang Wang, Zhimao Peng, Zhengyuan Xie, Fei Yang, Xialei Liu, Ming-Ming Cheng

This paper introduces textual information for the GCD task. By projecting visual features into the input space of the CLIP’s text encoder, it solves the problem that unlabeled data cannot be processed by the text encoder.
Through cross-modal distillation and multi-modal joint training, GET achieves state-of-the-art results on multiple benchmarks, providing GCD a multi-modal paradigm.

NeurIPS 2025

Enhancing Visual Grounding for GUI Agents via Self-Evolutionary Reinforcement Learning
Xinbin Yuan, Jian Zhang, Kaixin Li, Zhuoxuan Cai, Lujian Yao, Jie Chen, Enguang Wang, Qibin Hou, Jinwei Chen, Peng-Tao Jiang, Bo Li

This paper builds a GUI agent with an RL-based framework that incorporates three core strategies: (1) seed data curation to ensure high-quality training samples, (2) a dense policy gradient that provides continuous feedback based on prediction accuracy, and (3) a self-evolutionary reinforcement finetuning mechanism that iteratively refines the model using attention maps.
With 3k samples, the 7B-model achieves SOTA results on three grounding benchmarks.

IEEE TMM

This paper proposes PartGCD, including 1) Adaptive Part Decomposition, which automatically extracts class-specific semantic parts via Gaussian Mixture Models, and 2) Part Discrepancy Regularization, enforcing explicit separation between part features to amplify fine-grained local part distinctions.
Experiments demonstrate state-of-the-art performance across multiple fine-grained benchmarks while maintaining competitiveness on generic datasets.

View All Publications...

IJCAI 2024

This paper introduces a novel method using Nearest Neighbor Distance-aware Label Consistency sample selection. It creates class-consistent subsets for novel class sample clusters from the current GCD method, acting as “pseudo-labeled sets” to mitigate representation bias.
It proposes progressive supervised representation learning with selected samples to optimize the trade-off between quantity and purity in each subset.
The results demonstrate the superiority of the method in achieving improved performance in GCD tasks.

IEEE TMM

This paper introduces a loss sharpness penalty into GCD, which generates more robust pseudo labels for training samples by minimizing the worst-case sharpness.
It proposes a novel dynamic anchor selection strategy to further improve the performance by assigning hard pseudo-labels to the selected unknown classes anchors.
Experimental results show a state-of-the-art performance

IEEE TCSVT

This paper analyzes the limitations of existing SCOOD methods based on the clustering-based binary In-distribution Filtering strategy and proposes to use dual-threshold ternary sample assignment to select more reliable auxiliary ID and OOD samples for the SCOOD task.
It proposes a predictive sample assignment framework for the SCOOD task, which includes a threshold sample assignment mechanism, an ID/OOD concept contrastive representation learning loss, and an additional sample retraining process. Experiments show a state-of-the-art performance.

ECCV 2024

This paper proposes a new classifier pre-tuning method that learns a transformation from all old classifiers to generate new classifiers with previous knowledge before the formal training process. It optimizes the initialization of transformation matrices, striking a balance between stability and plasticity.
Results show that the method can be readily integrated into other approaches and significantly enhances their performance.

IEEE TCSVT

This paper points out that existing CCSSL methods overlook the issue of inter-task data comparisons, and proposes BGE to incorporate external data into training to address this gap. It proposes a sampling algorithm to sample external data that are relevant to tasks and sufficiently diverse, while also filtering out OOD data.
Experiments show that BGE can be seamlessly integrated into existing CCSSL methods and consistently yields significant improvement.

IEEE TMM

This paper highlights that the testing stages can help reduce forgetting and proposes a novel test-time semantic evolution method, RoSE, which is based on a tailored auxiliary task designed to restore forgotten knowledge. During test time, it computes the analytical solution rather than the gradient for updating the model, addressing the limitations of online training.
Extensive experiments show that RoSE outperforms most state-of-the-art NECIL methods across various scenarios.

CVMJ

This paper proposes a module to generate auxiliary dummy classes to learn richer prior knowledge for NCD tasks, so that the model can generate less noisy pseudo labels in unsupervised clustering. It also introduces a cluster-guided dummy classes synthesis strategy, the number of synthesized dummy classes can be significantly reduced and further improve the NCD performance.
The proposed method can outperform existing methods by a significant margin