Zhuang Fu

Zeyu Fu, Zhuang Fu, Chenzhuo Lu et al.

The extraction of contrast-filled vessels from X-ray coronary angiography (XCA) image sequence has important clinical significance for intuitively diagnosis and therapy. In this study, the XCA image sequence is regarded as a 3D tensor input, the vessel layer is regarded as a sparse tensor, and the background layer is regarded as a low-rank tensor. Using tensor nuclear norm (TNN) minimization, a novel method for vessel layer extraction based on tensor robust principal component analysis (TRPCA) is proposed. Furthermore, considering the irregular movement of vessels and the low-frequency dynamic disturbance of surrounding irrelevant tissues, the total variation (TV) regularized spatial-temporal constraint is introduced to smooth the foreground layer. Subsequently, for vessel layer images with uneven contrast distribution, a two-stage region growing (TSRG) method is utilized for vessel enhancement and segmentation. A global threshold method is used as the preprocessing to obtain main branches, and the Radon-Like features (RLF) filter is used to enhance and connect broken minor segments, the final binary vessel mask is constructed by combining the two intermediate results. The visibility of TV-TRPCA algorithm for foreground extraction is evaluated on clinical XCA image sequences and third-party dataset, which can effectively improve the performance of commonly used vessel segmentation algorithms. Based on TV-TRPCA, the accuracy of TSRG algorithm for vessel segmentation is further evaluated. Both qualitative and quantitative results validate the superiority of the proposed method over existing state-of-the-art approaches.

Haichao Wei, Yunxiang Ren, Zhoutong Fu et al.

Large Language Models (LLMs) demand significant computational resources, making it essential to enhance their capabilities without retraining from scratch. A key challenge in this domain is \textit{catastrophic forgetting} (CF), which hampers performance during Continuous Pre-training (CPT) and Continuous Supervised Fine-Tuning (CSFT). We propose \textbf{Control LLM}, a novel approach that leverages parallel pre-trained and expanded transformer blocks, aligning their hidden-states through interpolation strategies This method effectively preserves performance on existing tasks while seamlessly integrating new knowledge. Extensive experiments demonstrate the effectiveness of Control LLM in both CPT and CSFT. On Llama3.1-8B-Instruct, it achieves significant improvements in mathematical reasoning ($+14.4\%$ on Math-Hard) and coding performance ($+10\%$ on MBPP-PLUS). On Llama3.1-8B, it enhances multilingual capabilities ($+10.6\%$ on C-Eval, $+6.8\%$ on CMMLU, and $+30.2\%$ on CMMLU-0shot-CoT). It surpasses existing methods and achieves SOTA among open-source models tuned from the same base model, using substantially less data and compute. Crucially, these gains are realized while preserving strong original capabilities, with minimal degradation ($<4.3\% \text{on MMLU}$) compared to $>35\%$ in open-source Math and Coding models. This approach has been successfully deployed in LinkedIn's GenAI-powered job seeker and Ads unit products. To support further research, we release the training and evaluation code (https://github.com/linkedin/ControlLLM) along with models trained on public datasets (https://huggingface.co/ControlLLM) to the community.

Ziwen Guo, Zi Fang, Zhuang Fu

Three-dimensional ultrasound imaging is a critical technology widely used in medical diagnostics. However, traditional 3D ultrasound imaging methods have limitations such as fixed resolution, low storage efficiency, and insufficient contextual connectivity, leading to poor performance in handling complex artifacts and reflection characteristics. Recently, techniques based on NeRF (Neural Radiance Fields) have made significant progress in view synthesis and 3D reconstruction, but there remains a research gap in high-quality ultrasound imaging. To address these issues, we propose a new model, UlRe-NeRF, which combines implicit neural networks and explicit ultrasound volume rendering into an ultrasound neural rendering architecture. This model incorporates reflection direction parameterization and harmonic encoding, using a directional MLP module to generate view-dependent high-frequency reflection intensity estimates, and a spatial MLP module to produce the medium's physical property parameters. These parameters are used in the volume rendering process to accurately reproduce the propagation and reflection behavior of ultrasound waves in the medium. Experimental results demonstrate that the UlRe-NeRF model significantly enhances the realism and accuracy of high-fidelity ultrasound image reconstruction, especially in handling complex medium structures.

Junjia Liu, Huimin Zhang, Zhuang Fu et al.

The intelligent control of the traffic signal is critical to the optimization of transportation systems. To achieve global optimal traffic efficiency in large-scale road networks, recent works have focused on coordination among intersections, which have shown promising results. However, existing studies paid more attention to observations sharing among intersections (both explicit and implicit) and did not care about the consequences after decisions. In this paper, we design a multiagent coordination framework based on Deep Reinforcement Learning methods for traffic signal control, defined as γ-Reward that includes both original γ-Reward and γ-Attention-Reward. Specifically, we propose the Spatial Differentiation method for coordination which uses the temporal-spatial information in the replay buffer to amend the reward of each action. A concise theoretical analysis that proves the proposed model can converge to Nash equilibrium is given. By extending the idea of Markov Chain to the dimension of space-time, this truly decentralized coordination mechanism replaces the graph attention method and realizes the decoupling of the road network, which is more scalable and more in line with practice. The simulation results show that the proposed model remains a state-of-the-art performance even not use a centralized setting. Code is available in https://github.com/Skylark0924/Gamma Reward.

Junjia Liu, Jiaying Shou, Zhuang Fu et al.

Compared to rigid robots that are generally studied in reinforcement learning, the physical characteristics of some sophisticated robots such as soft or continuum robots are higher complicated. Moreover, recent reinforcement learning methods are data-inefficient and can not be directly deployed to the robot without simulation. In this paper, we propose an efficient reinforcement learning method based on inexplicit prior knowledge in response to such problems. We first corroborate the method by simulation and employed directly in the real world. By using our method, we can achieve active visual tracking and distance maintenance of a tendon-driven robot which will be critical in minimally invasive procedures. Codes are available at https://github.com/Skylark0924/TendonTrack.