Qi Yang

Jiaxing Zhang, Ruyi Gan, Junjie Wang et al.

Nowadays, foundation models become one of fundamental infrastructures in artificial intelligence, paving ways to the general intelligence. However, the reality presents two urgent challenges: existing foundation models are dominated by the English-language community; users are often given limited resources and thus cannot always use foundation models. To support the development of the Chinese-language community, we introduce an open-source project, called Fengshenbang, which leads by the research center for Cognitive Computing and Natural Language (CCNL). Our project has comprehensive capabilities, including large pre-trained models, user-friendly APIs, benchmarks, datasets, and others. We wrap all these in three sub-projects: the Fengshenbang Model, the Fengshen Framework, and the Fengshen Benchmark. An open-source roadmap, Fengshenbang, aims to re-evaluate the open-source community of Chinese pre-trained large-scale models, prompting the development of the entire Chinese large-scale model community. We also want to build a user-centered open-source ecosystem to allow individuals to access the desired models to match their computing resources. Furthermore, we invite companies, colleges, and research institutions to collaborate with us to build the large-scale open-source model-based ecosystem. We hope that this project will be the foundation of Chinese cognitive intelligence.

Xin Yu, Qi Yang, Yucheng Tang et al.

2D low-dose single-slice abdominal computed tomography (CT) slice enables direct measurements of body composition, which are critical to quantitatively characterizing health relationships on aging. However, longitudinal analysis of body composition changes using 2D abdominal slices is challenging due to positional variance between longitudinal slices acquired in different years. To reduce the positional variance, we extend the conditional generative models to our C-SliceGen that takes an arbitrary axial slice in the abdominal region as the condition and generates a defined vertebral level slice by estimating the structural changes in the latent space. Experiments on 1170 subjects from an in-house dataset and 50 subjects from BTCV MICCAI Challenge 2015 show that our model can generate high quality images in terms of realism and similarity. External experiments on 20 subjects from the Baltimore Longitudinal Study of Aging (BLSA) dataset that contains longitudinal single abdominal slices validate that our method can harmonize the slice positional variance in terms of muscle and visceral fat area. Our approach provides a promising direction of mapping slices from different vertebral levels to a target slice to reduce positional variance for single slice longitudinal analysis. The source code is available at: https://github.com/MASILab/C-SliceGen.

Qi Yang, Xin Yu, Ho Hin Lee et al.

Objective: Thigh muscle group segmentation is important for assessment of muscle anatomy, metabolic disease and aging. Many efforts have been put into quantifying muscle tissues with magnetic resonance (MR) imaging including manual annotation of individual muscles. However, leveraging publicly available annotations in MR images to achieve muscle group segmentation on single slice computed tomography (CT) thigh images is challenging. Method: We propose an unsupervised domain adaptation pipeline with self-training to transfer labels from 3D MR to single CT slice. First, we transform the image appearance from MR to CT with CycleGAN and feed the synthesized CT images to a segmenter simultaneously. Single CT slices are divided into hard and easy cohorts based on the entropy of pseudo labels inferenced by the segmenter. After refining easy cohort pseudo labels based on anatomical assumption, self-training with easy and hard splits is applied to fine tune the segmenter. Results: On 152 withheld single CT thigh images, the proposed pipeline achieved a mean Dice of 0.888(0.041) across all muscle groups including sartorius, hamstrings, quadriceps femoris and gracilis. muscles Conclusion: To our best knowledge, this is the first pipeline to achieve thigh imaging domain adaptation from MR to CT. The proposed pipeline is effective and robust in extracting muscle groups on 2D single slice CT thigh images.The container is available for public use at https://github.com/MASILab/DA_CT_muscle_seg

Xin Yu, Yucheng Tang, Qi Yang et al.

Whole brain segmentation with magnetic resonance imaging (MRI) enables the non-invasive measurement of brain regions, including total intracranial volume (TICV) and posterior fossa volume (PFV). Enhancing the existing whole brain segmentation methodology to incorporate intracranial measurements offers a heightened level of comprehensiveness in the analysis of brain structures. Despite its potential, the task of generalizing deep learning techniques for intracranial measurements faces data availability constraints due to limited manually annotated atlases encompassing whole brain and TICV/PFV labels. In this paper, we enhancing the hierarchical transformer UNesT for whole brain segmentation to achieve segmenting whole brain with 133 classes and TICV/PFV simultaneously. To address the problem of data scarcity, the model is first pretrained on 4859 T1-weighted (T1w) 3D volumes sourced from 8 different sites. These volumes are processed through a multi-atlas segmentation pipeline for label generation, while TICV/PFV labels are unavailable. Subsequently, the model is finetuned with 45 T1w 3D volumes from Open Access Series Imaging Studies (OASIS) where both 133 whole brain classes and TICV/PFV labels are available. We evaluate our method with Dice similarity coefficients(DSC). We show that our model is able to conduct precise TICV/PFV estimation while maintaining the 132 brain regions performance at a comparable level. Code and trained model are available at: https://github.com/MASILab/UNesT/tree/main/wholebrainSeg.

Xin Yu, Qi Yang, Yucheng Tang et al.

Two-dimensional single-slice abdominal computed tomography (CT) provides a detailed tissue map with high resolution allowing quantitative characterization of relationships between health conditions and aging. However, longitudinal analysis of body composition changes using these scans is difficult due to positional variation between slices acquired in different years, which leading to different organs/tissues captured. To address this issue, we propose C-SliceGen, which takes an arbitrary axial slice in the abdominal region as a condition and generates a pre-defined vertebral level slice by estimating structural changes in the latent space. Our experiments on 2608 volumetric CT data from two in-house datasets and 50 subjects from the 2015 Multi-Atlas Abdomen Labeling Challenge dataset (BTCV) Challenge demonstrate that our model can generate high-quality images that are realistic and similar. We further evaluate our method's capability to harmonize longitudinal positional variation on 1033 subjects from the Baltimore Longitudinal Study of Aging (BLSA) dataset, which contains longitudinal single abdominal slices, and confirmed that our method can harmonize the slice positional variance in terms of visceral fat area. This approach provides a promising direction for mapping slices from different vertebral levels to a target slice and reducing positional variance for single-slice longitudinal analysis. The source code is available at: https://github.com/MASILab/C-SliceGen.

Ho Hin Lee, Yucheng Tang, Riqiang Gao et al.

Non-contrast computed tomography (NCCT) is commonly acquired for lung cancer screening, assessment of general abdominal pain or suspected renal stones, trauma evaluation, and many other indications. However, the absence of contrast limits distinguishing organ in-between boundaries. In this paper, we propose a novel unsupervised approach that leverages pairwise contrast-enhanced CT (CECT) context to compute non-contrast segmentation without ground-truth label. Unlike generative adversarial approaches, we compute the pairwise morphological context with CECT to provide teacher guidance instead of generating fake anatomical context. Additionally, we further augment the intensity correlations in 'organ-specific' settings and increase the sensitivity to organ-aware boundary. We validate our approach on multi-organ segmentation with paired non-contrast & contrast-enhanced CT scans using five-fold cross-validation. Full external validations are performed on an independent non-contrast cohort for aorta segmentation. Compared with current abdominal organs segmentation state-of-the-art in fully supervised setting, our proposed pipeline achieves a significantly higher Dice by 3.98% (internal multi-organ annotated), and 8.00% (external aorta annotated) for abdominal organs segmentation. The code and pretrained models are publicly available at https://github.com/MASILab/ContrastMix.

Yujie Zhang, Qi Yang, Yifei Zhou et al.

The goal of objective point cloud quality assessment (PCQA) research is to develop quantitative metrics that measure point cloud quality in a perceptually consistent manner. Merging the research of cognitive science and intuition of the human visual system (HVS), in this paper, we evaluate the point cloud quality by measuring the complexity of transforming the distorted point cloud back to its reference, which in practice can be approximated by the code length of one point cloud when the other is given. For this purpose, we first make space segmentation for the reference and distorted point clouds based on a 3D Voronoi diagram to obtain a series of local patch pairs. Next, inspired by the predictive coding theory, we utilize a space-aware vector autoregressive (SA-VAR) model to encode the geometry and color channels of each reference patch with and without the distorted patch, respectively. Assuming that the residual errors follow the multi-variate Gaussian distributions, the self-complexity of the reference and transformational complexity between the reference and distorted samples are computed using covariance matrices. Additionally, the prediction terms generated by SA-VAR are introduced as one auxiliary feature to promote the final quality prediction. The effectiveness of the proposed transformational complexity based distortion metric (TCDM) is evaluated through extensive experiments conducted on five public point cloud quality assessment databases. The results demonstrate that TCDM achieves state-of-the-art (SOTA) performance, and further analysis confirms its robustness in various scenarios. The code is publicly available at https://github.com/zyj1318053/TCDM.

Yuke Xing, Qi Yang, Kaifa Yang et al.

In recent years, Neural Radiance Fields (NeRF) have demonstrated significant advantages in representing and synthesizing 3D scenes. Explicit NeRF models facilitate the practical NeRF applications with faster rendering speed, and also attract considerable attention in NeRF compression due to its huge storage cost. To address the challenge of the NeRF compression study, in this paper, we construct a new dataset, called Explicit-NeRF-QA. We use 22 3D objects with diverse geometries, textures, and material complexities to train four typical explicit NeRF models across five parameter levels. Lossy compression is introduced during the model generation, pivoting the selection of key parameters such as hash table size for InstantNGP and voxel grid resolution for Plenoxels. By rendering NeRF samples to processed video sequences (PVS), a large scale subjective experiment with lab environment is conducted to collect subjective scores from 21 viewers. The diversity of content, accuracy of mean opinion scores (MOS), and characteristics of NeRF distortion are comprehensively presented, establishing the heterogeneity of the proposed dataset. The state-of-the-art objective metrics are tested in the new dataset. Best Person correlation, which is around 0.85, is collected from the full-reference objective metric. All tested no-reference metrics report very poor results with 0.4 to 0.6 correlations, demonstrating the need for further development of more robust no-reference metrics. The dataset, including NeRF samples, source 3D objects, multiview images for NeRF generation, PVSs, MOS, is made publicly available at the following location: https://github.com/YukeXing/Explicit-NeRF-QA.

Ho Hin Lee, Yucheng Tang, Han Liu et al.

Recent studies have demonstrated the superior performance of introducing ``scan-wise" contrast labels into contrastive learning for multi-organ segmentation on multi-phase computed tomography (CT). However, such scan-wise labels are limited: (1) a coarse classification, which could not capture the fine-grained ``organ-wise" contrast variations across all organs; (2) the label (i.e., contrast phase) is typically manually provided, which is error-prone and may introduce manual biases of defining phases. In this paper, we propose a novel data-driven contrastive loss function that adapts the similar/dissimilar contrast relationship between samples in each minibatch at organ-level. Specifically, as variable levels of contrast exist between organs, we hypothesis that the contrast differences in the organ-level can bring additional context for defining representations in the latent space. An organ-wise contrast correlation matrix is computed with mean organ intensities under one-hot attention maps. The goal of adapting the organ-driven correlation matrix is to model variable levels of feature separability at different phases. We evaluate our proposed approach on multi-organ segmentation with both non-contrast CT (NCCT) datasets and the MICCAI 2015 BTCV Challenge contrast-enhance CT (CECT) datasets. Compared to the state-of-the-art approaches, our proposed contrastive loss yields a substantial and significant improvement of 1.41% (from 0.923 to 0.936, p-value$<$0.01) and 2.02% (from 0.891 to 0.910, p-value$<$0.01) on mean Dice scores across all organs with respect to NCCT and CECT cohorts. We further assess the trained model performance with the MICCAI 2021 FLARE Challenge CECT datasets and achieve a substantial improvement of mean Dice score from 0.927 to 0.934 (p-value$<$0.01). The code is available at: https://github.com/MASILab/DCC_CL

Wei Zhou, Qi Yang, Qiuping Jiang et al.

Objective quality assessment of 3D point clouds is essential for the development of immersive multimedia systems in real-world applications. Despite the success of perceptual quality evaluation for 2D images and videos, blind/no-reference metrics are still scarce for 3D point clouds with large-scale irregularly distributed 3D points. Therefore, in this paper, we propose an objective point cloud quality index with Structure Guided Resampling (SGR) to automatically evaluate the perceptually visual quality of 3D dense point clouds. The proposed SGR is a general-purpose blind quality assessment method without the assistance of any reference information. Specifically, considering that the human visual system (HVS) is highly sensitive to structure information, we first exploit the unique normal vectors of point clouds to execute regional pre-processing which consists of keypoint resampling and local region construction. Then, we extract three groups of quality-related features, including: 1) geometry density features; 2) color naturalness features; 3) angular consistency features. Both the cognitive peculiarities of the human brain and naturalness regularity are involved in the designed quality-aware features that can capture the most vital aspects of distorted 3D point clouds. Extensive experiments on several publicly available subjective point cloud quality databases validate that our proposed SGR can compete with state-of-the-art full-reference, reduced-reference, and no-reference quality assessment algorithms.

Ruyi Gan, Ziwei Wu, Renliang Sun et al.

Various large language models (LLMs) have been proposed in recent years, including closed- and open-source ones, continually setting new records on multiple benchmarks. However, the development of LLMs still faces several issues, such as high cost of training models from scratch, and continual pre-training leading to catastrophic forgetting, etc. Although many such issues are addressed along the line of research on LLMs, an important yet practical limitation is that many studies overly pursue enlarging model sizes without comprehensively analyzing and optimizing the use of pre-training data in their learning process, as well as appropriate organization and leveraging of such data in training LLMs under cost-effective settings. In this work, we propose Ziya2, a model with 13 billion parameters adopting LLaMA2 as the foundation model, and further pre-trained on 700 billion tokens, where we focus on pre-training techniques and use data-centric optimization to enhance the learning process of Ziya2 on different stages. We define three data attributes and firstly establish data-centric scaling laws to illustrate how different data impacts LLMs. Experiments show that Ziya2 significantly outperforms other models in multiple benchmarks especially with promising results compared to representative open-source ones. Ziya2 (Base) is released at https://huggingface.co/IDEA-CCNL/Ziya2-13B-Base and https://modelscope.cn/models/Fengshenbang/Ziya2-13B-Base/summary.

Nuo Chen, Hongguang Li, Junqing He et al.

The conversational machine reading comprehension (CMRC) task aims to answer questions in conversations, which has been a hot research topic in recent years because of its wide applications. However, existing CMRC benchmarks in which each conversation is assigned a static passage are inconsistent with real scenarios. Thus, model's comprehension ability towards real scenarios are hard to evaluate reasonably. To this end, we propose the first Chinese CMRC benchmark Orca and further provide zero-shot/few-shot settings to evaluate model's generalization ability towards diverse domains. We collect 831 hot-topic driven conversations with 4,742 turns in total. Each turn of a conversation is assigned with a response-related passage, aiming to evaluate model's comprehension ability more reasonably. The topics of conversations are collected from social media platform and cover 33 domains, trying to be consistent with real scenarios. Importantly, answers in Orca are all well-annotated natural responses rather than the specific spans or short phrase in previous datasets. Besides, we implement three strong baselines to tackle the challenge in Orca. The results indicate the great challenge of our CMRC benchmark. Our datatset and checkpoints are available at https://github.com/nuochenpku/Orca.

Yujie Zhang, Qi Yang, Yiling Xu et al.

Full-reference point cloud quality assessment (FR-PCQA) aims to infer the quality of distorted point clouds with available references. Most of the existing FR-PCQA metrics ignore the fact that the human visual system (HVS) dynamically tackles visual information according to different distortion levels (i.e., distortion detection for high-quality samples and appearance perception for low-quality samples) and measure point cloud quality using unified features. To bridge the gap, in this paper, we propose a perception-guided hybrid metric (PHM) that adaptively leverages two visual strategies with respect to distortion degree to predict point cloud quality: to measure visible difference in high-quality samples, PHM takes into account the masking effect and employs texture complexity as an effective compensatory factor for absolute difference; on the other hand, PHM leverages spectral graph theory to evaluate appearance degradation in low-quality samples. Variations in geometric signals on graphs and changes in the spectral graph wavelet coefficients are utilized to characterize geometry and texture appearance degradation, respectively. Finally, the results obtained from the two components are combined in a non-linear method to produce an overall quality score of the tested point cloud. The results of the experiment on five independent databases show that PHM achieves state-of-the-art (SOTA) performance and offers significant performance improvement in multiple distortion environments. The code is publicly available at https://github.com/zhangyujie-1998/PHM.

Mao Zheng, Zheng Li, Tao Chen et al.

Hy-MT2 is a family of fast-thinking multilingual translation models designed for complex real-world scenarios. It includes three model sizes: 1.8B, 7B, and 30B-A3B (MoE), all of which support translation among 33 languages and effectively follow translation instructions in multiple languages. For on-device deployment, with AngelSlim 1.25-bit extreme quantization, the 1.8B model requires only 440 MB of storage and improves inference speed by 1.5x. Multi-dimensional evaluations show that Hy-MT2 delivers outstanding performance across general, real-world business, domain-specific, and instruction-following translation tasks. The 7B and 30B models outperform open-source models such as DeepSeek-V4-Pro and Kimi K2.6 in fast-thinking mode, while the lightweight 1.8B model also surpasses mainstream commercial APIs from providers such as Microsoft and Doubao overall.

Yujie Zhang, Qi Yang, Ziyu Shan et al.

Recent years have witnessed the success of the deep learning-based technique in research of no-reference point cloud quality assessment (NR-PCQA). For a more accurate quality prediction, many previous studies have attempted to capture global and local features in a bottom-up manner, but ignored the interaction and promotion between them. To solve this problem, we propose a novel asynchronous feedback quality prediction network (AFQ-Net). Motivated by human visual perception mechanisms, AFQ-Net employs a dual-branch structure to deal with global and local features, simulating the left and right hemispheres of the human brain, and constructs a feedback module between them. Specifically, the input point clouds are first fed into a transformer-based global encoder to generate the attention maps that highlight these semantically rich regions, followed by being merged into the global feature. Then, we utilize the generated attention maps to perform dynamic convolution for different semantic regions and obtain the local feature. Finally, a coarse-to-fine strategy is adopted to merge the two features into the final quality score. We conduct comprehensive experiments on three datasets and achieve superior performance over the state-of-the-art approaches on all of these datasets. The code will be available at The code will be available at https://github.com/zhangyujie-1998/AFQ-Net.

Xin Yu, Yucheng Tang, Qi Yang et al.

Metabolic health is increasingly implicated as a risk factor across conditions from cardiology to neurology, and efficiency assessment of body composition is critical to quantitatively characterizing these relationships. 2D low dose single slice computed tomography (CT) provides a high resolution, quantitative tissue map, albeit with a limited field of view. Although numerous potential analyses have been proposed in quantifying image context, there has been no comprehensive study for low-dose single slice CT longitudinal variability with automated segmentation. We studied a total of 1816 slices from 1469 subjects of Baltimore Longitudinal Study on Aging (BLSA) abdominal dataset using supervised deep learning-based segmentation and unsupervised clustering method. 300 out of 1469 subjects that have two year gap in their first two scans were pick out to evaluate longitudinal variability with measurements including intraclass correlation coefficient (ICC) and coefficient of variation (CV) in terms of tissues/organs size and mean intensity. We showed that our segmentation methods are stable in longitudinal settings with Dice ranged from 0.821 to 0.962 for thirteen target abdominal tissues structures. We observed high variability in most organ with ICC<0.5, low variability in the area of muscle, abdominal wall, fat and body mask with average ICC>0.8. We found that the variability in organ is highly related to the cross-sectional position of the 2D slice. Our efforts pave quantitative exploration and quality control to reduce uncertainties in longitudinal analysis.

Tencent Hunyuan Team, Ao Liu, Botong Zhou et al. · tencent-ai

As Large Language Models (LLMs) rapidly advance, we introduce Hunyuan-TurboS, a novel large hybrid Transformer-Mamba Mixture of Experts (MoE) model. It synergistically combines Mamba's long-sequence processing efficiency with Transformer's superior contextual understanding. Hunyuan-TurboS features an adaptive long-short chain-of-thought (CoT) mechanism, dynamically switching between rapid responses for simple queries and deep "thinking" modes for complex problems, optimizing computational resources. Architecturally, this 56B activated (560B total) parameter model employs 128 layers (Mamba2, Attention, FFN) with an innovative AMF/MF block pattern. Faster Mamba2 ensures linear complexity, Grouped-Query Attention minimizes KV cache, and FFNs use an MoE structure. Pre-trained on 16T high-quality tokens, it supports a 256K context length and is the first industry-deployed large-scale Mamba model. Our comprehensive post-training strategy enhances capabilities via Supervised Fine-Tuning (3M instructions), a novel Adaptive Long-short CoT Fusion method, Multi-round Deliberation Learning for iterative improvement, and a two-stage Large-scale Reinforcement Learning process targeting STEM and general instruction-following. Evaluations show strong performance: overall top 7 rank on LMSYS Chatbot Arena with a score of 1356, outperforming leading models like Gemini-2.0-Flash-001 (1352) and o4-mini-2025-04-16 (1345). TurboS also achieves an average of 77.9% across 23 automated benchmarks. Hunyuan-TurboS balances high performance and efficiency, offering substantial capabilities at lower inference costs than many reasoning models, establishing a new paradigm for efficient large-scale pre-trained models.

Qi Yang, Kaifa Yang, Yuke Xing et al.

To fill the gap of traditional GS compression method, in this paper, we first propose a simple and effective GS data compression anchor called Graph-based GS Compression (GGSC). GGSC is inspired by graph signal processing theory and uses two branches to compress the primitive center and attributes. We split the whole GS sample via KDTree and clip the high-frequency components after the graph Fourier transform. Followed by quantization, G-PCC and adaptive arithmetic coding are used to compress the primitive center and attribute residual matrix to generate the bitrate file. GGSS is the first work to explore traditional GS compression, with advantages that can reveal the GS distortion characteristics corresponding to typical compression operation, such as high-frequency clipping and quantization. Second, based on GGSC, we create a GS Quality Assessment dataset (GSQA) with 120 samples. A subjective experiment is conducted in a laboratory environment to collect subjective scores after rendering GS into Processed Video Sequences (PVS). We analyze the characteristics of different GS distortions based on Mean Opinion Scores (MOS), demonstrating the sensitivity of different attributes distortion to visual quality. The GGSC code and the dataset, including GS samples, MOS, and PVS, are made publicly available at https://github.com/Qi-Yangsjtu/GGSC.

Ho Hin Lee, Quan Liu, Shunxing Bao et al.

With the inspiration of vision transformers, the concept of depth-wise convolution revisits to provide a large Effective Receptive Field (ERF) using Large Kernel (LK) sizes for medical image segmentation. However, the segmentation performance might be saturated and even degraded as the kernel sizes scaled up (e.g., $21\times 21\times 21$) in a Convolutional Neural Network (CNN). We hypothesize that convolution with LK sizes is limited to maintain an optimal convergence for locality learning. While Structural Re-parameterization (SR) enhances the local convergence with small kernels in parallel, optimal small kernel branches may hinder the computational efficiency for training. In this work, we propose RepUX-Net, a pure CNN architecture with a simple large kernel block design, which competes favorably with current network state-of-the-art (SOTA) (e.g., 3D UX-Net, SwinUNETR) using 6 challenging public datasets. We derive an equivalency between kernel re-parameterization and the branch-wise variation in kernel convergence. Inspired by the spatial frequency in the human visual system, we extend to vary the kernel convergence into element-wise setting and model the spatial frequency as a Bayesian prior to re-parameterize convolutional weights during training. Specifically, a reciprocal function is leveraged to estimate a frequency-weighted value, which rescales the corresponding kernel element for stochastic gradient descent. From the experimental results, RepUX-Net consistently outperforms 3D SOTA benchmarks with internal validation (FLARE: 0.929 to 0.944), external validation (MSD: 0.901 to 0.932, KiTS: 0.815 to 0.847, LiTS: 0.933 to 0.949, TCIA: 0.736 to 0.779) and transfer learning (AMOS: 0.880 to 0.911) scenarios in Dice Score.

Xin Yu, Yucheng Tang, Yinchi Zhou et al.

Efficiently quantifying renal structures can provide distinct spatial context and facilitate biomarker discovery for kidney morphology. However, the development and evaluation of the transformer model to segment the renal cortex, medulla, and collecting system remains challenging due to data inefficiency. Inspired by the hierarchical structures in vision transformer, we propose a novel method using a 3D block aggregation transformer for segmenting kidney components on contrast-enhanced CT scans. We construct the first cohort of renal substructures segmentation dataset with 116 subjects under institutional review board (IRB) approval. Our method yields the state-of-the-art performance (Dice of 0.8467) against the baseline approach of 0.8308 with the data-efficient design. The Pearson R achieves 0.9891 between the proposed method and manual standards and indicates the strong correlation and reproducibility for volumetric analysis. We extend the proposed method to the public KiTS dataset, the method leads to improved accuracy compared to transformer-based approaches. We show that the 3D block aggregation transformer can achieve local communication between sequence representations without modifying self-attention, and it can serve as an accurate and efficient quantification tool for characterizing renal structures.

Qi Yang, Geert Van Der Auwera, Zhu Li

Current video-based GS compression methods rely on using Parallel Linear Assignment Sorting (PLAS) to convert 3D GS into smooth 2D maps, which are computationally expensive and time-consuming, limiting the application of GS on lightweight devices. In this paper, we propose a Lightweight 3D Gaussian Splatting (GS) Compression method based on Video codec (LGSCV). First, a two-stage Morton scan is proposed to generate blockwise 2D maps that are friendly for canonical video codecs in which the coding units (CU) are square blocks. A 3D Morton scan is used to permute GS primitives, followed by a 2D Morton scan to map the ordered GS primitives to 2D maps in a blockwise style. However, although the blockwise 2D maps report close performance to the PLAS map in high-bitrate regions, they show a quality collapse at medium-to-low bitrates. Therefore, a principal component analysis (PCA) is used to reduce the dimensionality of spherical harmonics (SH), and a MiniPLAS, which is flexible and fast, is designed to permute the primitives within certain block sizes. Incorporating SH PCA and MiniPLAS leads to a significant gain in rate-distortion (RD) performance, especially at medium and low bitrates. MiniPLAS can also guide the setting of the codec CU size configuration and significantly reduce encoding time. Experimental results on the MPEG dataset demonstrate that the proposed LGSCV achieves over 20% RD gain compared with state-of-the-art methods, while reducing 2D map generation time to approximately 1 second and cutting encoding time by 50%. The code is available at https://github.com/Qi-Yangsjtu/LGSCV .

Qi Yang, Sergey Nikolenko, Alfred Huang et al.

Social media marketing plays a vital role in promoting brand and product values to wide audiences. In order to boost their advertising revenues, global media buying platforms such as Facebook Ads constantly reduce the reach of branded organic posts, pushing brands to spend more on paid media ads. In order to run organic and paid social media marketing efficiently, it is necessary to understand the audience, tailoring the content to fit their interests and online behaviours, which is impossible to do manually at a large scale. At the same time, various personality type categorization schemes such as the Myers-Briggs Personality Type indicator make it possible to reveal the dependencies between personality traits and user content preferences on a wider scale by categorizing audience behaviours in a unified and structured manner. This problem is yet to be studied in depth by the research community, while the level of impact of different personality traits on content recommendation accuracy has not been widely utilised and comprehensively evaluated so far. Specifically, in this work we investigate the impact of human personality traits on the content recommendation model by applying a novel personality-driven multi-view content recommender system called Personality Content Marketing Recommender Engine, or PersiC. Our experimental results and real-world case study demonstrate not just PersiC's ability to perform efficient human personality-driven multi-view content recommendation, but also allow for actionable digital ad strategy recommendations, which when deployed are able to improve digital advertising efficiency by over 420% as compared to the original human-guided approach.

Ziyu Shan, Qi Yang, Rui Ye et al.

With the rapid development of 3D vision, point cloud has become an increasingly popular 3D visual media content. Due to the irregular structure, point cloud has posed novel challenges to the related research, such as compression, transmission, rendering and quality assessment. In these latest researches, point cloud quality assessment (PCQA) has attracted wide attention due to its significant role in guiding practical applications, especially in many cases where the reference point cloud is unavailable. However, current no-reference metrics which based on prevalent deep neural network have apparent disadvantages. For example, to adapt to the irregular structure of point cloud, they require preprocessing such as voxelization and projection that introduce extra distortions, and the applied grid-kernel networks, such as Convolutional Neural Networks, fail to extract effective distortion-related features. Besides, they rarely consider the various distortion patterns and the philosophy that PCQA should exhibit shifting, scaling, and rotational invariance. In this paper, we propose a novel no-reference PCQA metric named the Graph convolutional PCQA network (GPA-Net). To extract effective features for PCQA, we propose a new graph convolution kernel, i.e., GPAConv, which attentively captures the perturbation of structure and texture. Then, we propose the multi-task framework consisting of one main task (quality regression) and two auxiliary tasks (distortion type and degree predictions). Finally, we propose a coordinate normalization module to stabilize the results of GPAConv under shift, scale and rotation transformations. Experimental results on two independent databases show that GPA-Net achieves the best performance compared to the state-of-the-art no-reference PCQA metrics, even better than some full-reference metrics in some cases.

Shengyuan Liu, Liuxin Bao, Qi Yang et al.

Medical image segmentation is evolving from task-specific models toward generalizable frameworks. Recent research leverages Multi-modal Large Language Models (MLLMs) as autonomous agents, employing reinforcement learning with verifiable reward (RLVR) to orchestrate specialized tools like the Segment Anything Model (SAM). However, these approaches often rely on single-turn, rigid interaction strategies and lack process-level supervision during training, which hinders their ability to fully exploit the dynamic potential of interactive tools and leads to redundant actions. To bridge this gap, we propose MedSAM-Agent, a framework that reformulates interactive segmentation as a multi-step autonomous decision-making process. First, we introduce a hybrid prompting strategy for expert-curated trajectory generation, enabling the model to internalize human-like decision heuristics and adaptive refinement strategies. Furthermore, we develop a two-stage training pipeline that integrates multi-turn, end-to-end outcome verification with a clinical-fidelity process reward design to promote interaction parsimony and decision efficiency. Extensive experiments across 6 medical modalities and 21 datasets demonstrate that MedSAM-Agent achieves state-of-the-art performance, effectively unifying autonomous medical reasoning with robust, iterative optimization. Code is available \href{https://github.com/CUHK-AIM-Group/MedSAM-Agent}{here}.

Qi Yang, Joel Jung, Timon Deschamps et al.

Dynamic colored meshes (DCM) are widely used in various applications; however, these meshes may undergo different processes, such as compression or transmission, which can distort them and degrade their quality. To facilitate the development of objective metrics for DCMs and study the influence of typical distortions on their perception, we create the Tencent - dynamic colored mesh database (TDMD) containing eight reference DCM objects with six typical distortions. Using processed video sequences (PVS) derived from the DCM, we have conducted a large-scale subjective experiment that resulted in 303 distorted DCM samples with mean opinion scores, making the TDMD the largest available DCM database to our knowledge. This database enabled us to study the impact of different types of distortion on human perception and offer recommendations for DCM compression and related tasks. Additionally, we have evaluated three types of state-of-the-art objective metrics on the TDMD, including image-based, point-based, and video-based metrics, on the TDMD. Our experimental results highlight the strengths and weaknesses of each metric, and we provide suggestions about the selection of metrics in practical DCM applications. The TDMD will be made publicly available at the following location: https://multimedia.tencent.com/resources/tdmd.

Zili Wang, Qi Yang, Linsu Shi et al.

Recently, transformer-based models have demonstrated remarkable performance on audio-visual segmentation (AVS) tasks. However, their expensive computational cost makes real-time inference impractical. By characterizing attention maps of the network, we identify two key obstacles in AVS models: 1) attention dissipation, corresponding to the over-concentrated attention weights by Softmax within restricted frames, and 2) inefficient, burdensome transformer decoder, caused by narrow focus patterns in early stages. In this paper, we introduce AVESFormer, the first real-time Audio-Visual Efficient Segmentation transformer that achieves fast, efficient and light-weight simultaneously. Our model leverages an efficient prompt query generator to correct the behaviour of cross-attention. Additionally, we propose ELF decoder to bring greater efficiency by facilitating convolutions suitable for local features to reduce computational burdens. Extensive experiments demonstrate that our AVESFormer significantly enhances model performance, achieving 79.9% on S4, 57.9% on MS3 and 31.2% on AVSS, outperforming previous state-of-the-art and achieving an excellent trade-off between performance and speed. Code can be found at https://github.com/MarkXCloud/AVESFormer.git.

Qi Yang, Shreya Ravikumar, Fynn Schmitt-Ulms et al.

We present an automatic large language model (LLM) conversion approach that produces uncertainty-aware LLMs capable of estimating uncertainty with every prediction. Our approach is model- and data-agnostic, is computationally-efficient, and does not rely on external models or systems. We evaluate converted models on the selective question answering setting -- to answer as many questions as possible while maintaining a given accuracy, forgoing providing predictions when necessary. As part of our results, we test BERT and Llama 2 model variants on the SQuAD extractive QA task and the TruthfulQA generative QA task. We show that using the uncertainty estimates provided by our approach to selectively answer questions leads to significantly higher accuracy over directly using model probabilities.

Chenyu Gao, Michael E. Kim, Ho Hin Lee et al.

Imaging findings inconsistent with those expected at specific chronological age ranges may serve as early indicators of neurological disorders and increased mortality risk. Estimation of chronological age, and deviations from expected results, from structural MRI data has become an important task for developing biomarkers that are sensitive to such deviations. Complementary to structural analysis, diffusion tensor imaging (DTI) has proven effective in identifying age-related microstructural changes within the brain white matter, thereby presenting itself as a promising additional modality for brain age prediction. Although early studies have sought to harness DTI's advantages for age estimation, there is no evidence that the success of this prediction is owed to the unique microstructural and diffusivity features that DTI provides, rather than the macrostructural features that are also available in DTI data. Therefore, we seek to develop white-matter-specific age estimation to capture deviations from normal white matter aging. Specifically, we deliberately disregard the macrostructural information when predicting age from DTI scalar images, using two distinct methods. The first method relies on extracting only microstructural features from regions of interest. The second applies 3D residual neural networks (ResNets) to learn features directly from the images, which are non-linearly registered and warped to a template to minimize macrostructural variations. When tested on unseen data, the first method yields mean absolute error (MAE) of 6.11 years for cognitively normal participants and MAE of 6.62 years for cognitively impaired participants, while the second method achieves MAE of 4.69 years for cognitively normal participants and MAE of 4.96 years for cognitively impaired participants. We find that the ResNet model captures subtler, non-macrostructural features for brain age prediction.

Ho Hin Lee, Quan Liu, Qi Yang et al.

The application of 3D ViTs to medical image segmentation has seen remarkable strides, somewhat overshadowing the budding advancements in Convolutional Neural Network (CNN)-based models. Large kernel depthwise convolution has emerged as a promising technique, showcasing capabilities akin to hierarchical transformers and facilitating an expansive effective receptive field (ERF) vital for dense predictions. Despite this, existing core operators, ranging from global-local attention to large kernel convolution, exhibit inherent trade-offs and limitations (e.g., global-local range trade-off, aggregating attentional features). We hypothesize that deformable convolution can be an exploratory alternative to combine all advantages from the previous operators, providing long-range dependency, adaptive spatial aggregation and computational efficiency as a foundation backbone. In this work, we introduce 3D DeformUX-Net, a pioneering volumetric CNN model that adeptly navigates the shortcomings traditionally associated with ViTs and large kernel convolution. Specifically, we revisit volumetric deformable convolution in depth-wise setting to adapt long-range dependency with computational efficiency. Inspired by the concepts of structural re-parameterization for convolution kernel weights, we further generate the deformable tri-planar offsets by adapting a parallel branch (starting from $1\times1\times1$ convolution), providing adaptive spatial aggregation across all channels. Our empirical evaluations reveal that the 3D DeformUX-Net consistently outperforms existing state-of-the-art ViTs and large kernel convolution models across four challenging public datasets, spanning various scales from organs (KiTS: 0.680 to 0.720, MSD Pancreas: 0.676 to 0.717, AMOS: 0.871 to 0.902) to vessels (e.g., MSD hepatic vessels: 0.635 to 0.671) in mean Dice.

Yinchi Zhou, Ho Hin Lee, Yucheng Tang et al.

With the substantial diversity in population demographics, such as differences in age and body composition, the volumetric morphology of pancreas varies greatly, resulting in distinctive variations in shape and appearance. Such variations increase the difficulty at generalizing population-wide pancreas features. A volumetric spatial reference is needed to adapt the morphological variability for organ-specific analysis. Here, we proposed a high-resolution computed tomography (CT) atlas framework specifically optimized for the pancreas organ across multi-contrast CT. We introduce a deep learning-based pre-processing technique to extract the abdominal region of interests (ROIs) and leverage a hierarchical registration pipeline to align the pancreas anatomy across populations. Briefly, DEEDs affine and non-rigid registration are performed to transfer patient abdominal volumes to a fixed high-resolution atlas template. To generate and evaluate the pancreas atlas template, multi-contrast modality CT scans of 443 subjects (without reported history of pancreatic disease, age: 15-50 years old) are processed. Comparing with different registration state-of-the-art tools, the combination of DEEDs affine and non-rigid registration achieves the best performance for the pancreas label transfer across all contrast phases. We further perform external evaluation with another research cohort of 100 de-identified portal venous scans with 13 organs labeled, having the best label transfer performance of 0.504 Dice score in unsupervised setting. The qualitative representation (e.g., average mapping) of each phase creates a clear boundary of pancreas and its distinctive contrast appearance. The deformation surface renderings across scales (e.g., small to large volume) further illustrate the generalizability of the proposed atlas template.

Lin Chen, Bolin Ni, Qi Yang et al.

Despite the remarkable capabilities of Multimodal Large Language Models (MLLMs), they still suffer from visual fading in long-context scenarios. Specifically, the attention to visual tokens diminishes as the text sequence lengthens, leading to text generation detached from visual constraints. We attribute this degradation to the inherent inductive bias of Multimodal RoPE, which penalizes inter-modal attention as the distance between visual and text tokens increases. To address this, we propose inter-modal Distance Invariant Position Encoding (DIPE), a simple but effective mechanism that disentangles position encoding based on modality interactions. DIPE retains the natural relative positioning for intra-modal interactions to preserve local structure, while enforcing an anchored perceptual proximity for inter-modal interactions. This strategy effectively mitigates the inter-modal distance-based penalty, ensuring that visual signals remain perceptually consistent regardless of the context length. Experimental results demonstrate that by integrating DIPE with Multimodal RoPE, the model maintains stable visual grounding in long-context scenarios, significantly alleviating visual fading while preserving performance on standard short-context benchmarks. Code is available at https://github.com/lchen1019/DIPE.

Bingyang Cui, Qi Yang, Kaifa Yang et al.

In recent years, static meshes with texture maps have become one of the most prevalent digital representations of 3D shapes in various applications, such as animation, gaming, medical imaging, and cultural heritage applications. However, little research has been done on the quality assessment of textured meshes, which hinders the development of quality-oriented applications, such as mesh compression and enhancement. In this paper, we create a large-scale textured mesh quality assessment database, namely SJTU-TMQA, which includes 21 reference meshes and 945 distorted samples. The meshes are rendered into processed video sequences and then conduct subjective experiments to obtain mean opinion scores (MOS). The diversity of content and accuracy of MOS has been shown to validate its heterogeneity and reliability. The impact of various types of distortion on human perception is demonstrated. 13 state-of-the-art objective metrics are evaluated on SJTU-TMQA. The results report the highest correlation of around 0.6, indicating the need for more effective objective metrics. The SJTU-TMQA is available at https://ccccby.github.io

Zhengyu Wang, Yujie Zhang, Qi Yang et al.

Point cloud quality assessment (PCQA) has become an appealing research field in recent days. Considering the importance of saliency detection in quality assessment, we propose an effective full-reference PCQA metric which makes the first attempt to utilize the saliency information to facilitate quality prediction, called point cloud quality assessment using 3D saliency maps (PQSM). Specifically, we first propose a projection-based point cloud saliency map generation method, in which depth information is introduced to better reflect the geometric characteristics of point clouds. Then, we construct point cloud local neighborhoods to derive three structural descriptors to indicate the geometry, color and saliency discrepancies. Finally, a saliency-based pooling strategy is proposed to generate the final quality score. Extensive experiments are performed on four independent PCQA databases. The results demonstrate that the proposed PQSM shows competitive performances compared to multiple state-of-the-art PCQA metrics.

Qi Yang, Joel Jung, Haiqiang Wang et al.

Static meshes with texture map are widely used in modern industrial and manufacturing sectors, attracting considerable attention in the mesh compression community due to its huge amount of data. To facilitate the study of static mesh compression algorithm and objective quality metric, we create the Tencent - Static Mesh Dataset (TSMD) containing 42 reference meshes with rich visual characteristics. 210 distorted samples are generated by the lossy compression scheme developed for the Call for Proposals on polygonal static mesh coding, released on June 23 by the Alliance for Open Media Volumetric Visual Media group. Using processed video sequences, a large-scale, crowdsourcing-based, subjective experiment was conducted to collect subjective scores from 74 viewers. The dataset undergoes analysis to validate its sample diversity and Mean Opinion Scores (MOS) accuracy, establishing its heterogeneous nature and reliability. State-of-the-art objective metrics are evaluated on the new dataset. Pearson and Spearman correlations around 0.75 are reported, deviating from results typically observed on less heterogeneous datasets, demonstrating the need for further development of more robust metrics. The TSMD, including meshes, PVSs, bitstreams, and MOS, is made publicly available at the following location: https://multimedia.tencent.com/resources/tsmd.

Kaifa Yang, Qi Yang, Yiling Xu et al.

3D Gaussian Splatting (3DGS) has emerged as a leading technology for high-quality 3D scene reconstruction. However, the iterative refinement and densification process leads to the generation of a large number of primitives, each contributing to the reconstruction to a substantially different extent. Estimating primitive importance is thus crucial, both for removing redundancy during reconstruction and for enabling efficient compression and transmission. Existing methods typically rely on rendering-based analyses, where each primitive is evaluated through its contribution across multiple camera viewpoints. However, such methods are sensitive to the number and selection of views, rely on specialized differentiable rasterizers, and have long calculation times that grow linearly with view count, making them difficult to integrate as plug-and-play modules and limiting scalability and generalization. To address these issues, we propose RAP, a fast feedforward rendering-free attribute-guided method for efficient importance score prediction in 3DGS. RAP infers primitive significance directly from intrinsic Gaussian attributes and local neighborhood statistics, avoiding rendering-based or visibility-dependent computations. A compact MLP predicts per-primitive importance scores using rendering loss, pruning-aware loss, and significance distribution regularization. After training on a small set of scenes, RAP generalizes effectively to unseen data and can be seamlessly integrated into reconstruction, compression, and transmission pipelines. Our code is publicly available at https://github.com/yyyykf/RAP.

Qi Yang, Xiangyao Ma, Xiao Wang et al.

As global cross-lingual communication intensifies, language barriers in visually rich documents such as PDFs remain a practical bottleneck. Existing document translation pipelines face a tension between linguistic processing and layout preservation: text-oriented Computer-Assisted Translation (CAT) systems often discard structural metadata, while document parsers focus on extraction and do not support faithful re-rendering after translation. We introduce BabelDOC, an Intermediate Representation (IR)-based framework for layout-preserving PDF translation. BabelDOC decouples visual layout metadata from semantic content, enabling document-level translation operations such as terminology extraction, cross-page context handling, glossary-constrained generation, and formula placeholdering. The translated content is then re-anchored to the original layout through an adaptive typesetting engine. Experiments on a curated 200-page benchmark, together with human evaluation and multimodal LLM-as-a-judge evaluation, show that BabelDOC improves layout fidelity, visual aesthetics, and terminology consistency over representative baselines, while maintaining competitive translation precision. The open-source toolkit and its interactive downstream applications are publicly available and have attracted over 8.4K GitHub stars and 17 contributors at the time of writing. A demonstration video is also available.

Zehong Wang, Xiaolong Han, Qi Yang et al.

Molecular representations fundamentally shape how machine learning systems reason about molecular structure and physical properties. Most existing approaches adopt a discrete pipeline: molecules are encoded as sequences, graphs, or point clouds, mapped to fixed-dimensional embeddings, and then used for task-specific prediction. This paradigm treats molecules as discrete objects, despite their intrinsically continuous and field-like physical nature. We argue that molecular learning can instead be formulated as learning in function space. Specifically, we model each molecule as a continuous function over three-dimensional (3D) space and treat this molecular field as the primary object of representation. From this perspective, conventional molecular representations arise as particular sampling schemes of an underlying continuous object. We instantiate this formulation with MolField, a hyper-network-based framework that learns distributions over molecular fields. To ensure physical consistency, these functions are defined over canonicalized coordinates, yielding invariance to global SE(3) transformations. To enable learning directly over functions, we introduce a structured weight tokenization and train a sequence-based hyper-network to model a shared prior over molecular fields. We evaluate MolField on molecular dynamics and property prediction. Our results show that treating molecules as continuous functions fundamentally changes how molecular representations generalize across tasks and yields downstream behavior that is stable to how molecules are discretized or queried.

Zhongwei Yu, Rasul Tutunov, Alexandre Max Maraval et al.

Traditional scientific discovery relies on an iterative hypothesise-experiment-refine cycle that has driven progress for centuries, but its intuitive, ad-hoc implementation often wastes resources, yields inefficient designs, and misses critical insights. This tutorial presents Bayesian Optimisation (BO), a principled probability-driven framework that formalises and automates this core scientific cycle. BO uses surrogate models (e.g., Gaussian processes) to model empirical observations as evolving hypotheses, and acquisition functions to guide experiment selection, balancing exploitation of known knowledge and exploration of uncharted domains to eliminate guesswork and manual trial-and-error. We first frame scientific discovery as an optimisation problem, then unpack BO's core components, end-to-end workflows, and real-world efficacy via case studies in catalysis, materials science, organic synthesis, and molecule discovery. We also cover critical technical extensions for scientific applications, including batched experimentation, heteroscedasticity, contextual optimisation, and human-in-the-loop integration. Tailored for a broad audience, this tutorial bridges AI advances in BO with practical natural science applications, offering tiered content to empower cross-disciplinary researchers to design more efficient experiments and accelerate principled scientific discovery.

Xin Yu, Qi Yang, Yinchi Zhou et al.

Transformer-based models, capable of learning better global dependencies, have recently demonstrated exceptional representation learning capabilities in computer vision and medical image analysis. Transformer reformats the image into separate patches and realizes global communication via the self-attention mechanism. However, positional information between patches is hard to preserve in such 1D sequences, and loss of it can lead to sub-optimal performance when dealing with large amounts of heterogeneous tissues of various sizes in 3D medical image segmentation. Additionally, current methods are not robust and efficient for heavy-duty medical segmentation tasks such as predicting a large number of tissue classes or modeling globally inter-connected tissue structures. To address such challenges and inspired by the nested hierarchical structures in vision transformer, we proposed a novel 3D medical image segmentation method (UNesT), employing a simplified and faster-converging transformer encoder design that achieves local communication among spatially adjacent patch sequences by aggregating them hierarchically. We extensively validate our method on multiple challenging datasets, consisting of multiple modalities, anatomies, and a wide range of tissue classes, including 133 structures in the brain, 14 organs in the abdomen, 4 hierarchical components in the kidneys, inter-connected kidney tumors and brain tumors. We show that UNesT consistently achieves state-of-the-art performance and evaluate its generalizability and data efficiency. Particularly, the model achieves whole brain segmentation task complete ROI with 133 tissue classes in a single network, outperforming prior state-of-the-art method SLANT27 ensembled with 27 networks.

Qi Yang, Joel Jung, Xiaozhong Xu et al.

Static meshes with texture maps have attracted considerable attention in both industrial manufacturing and academic research, leading to an urgent requirement for effective and robust objective quality evaluation. However, current model-based static mesh quality metrics have obvious limitations: most of them only consider geometry information, while color information is ignored, and they have strict constraints for the meshes' geometrical topology. Other metrics, such as image-based and point-based metrics, are easily influenced by the prepossessing algorithms, e.g., projection and sampling, hampering their ability to perform at their best. In this paper, we propose Geodesic Patch Similarity (GeodesicPSIM), a novel model-based metric to accurately predict human perception quality for static meshes. After selecting a group keypoints, 1-hop geodesic patches are constructed based on both the reference and distorted meshes cleaned by an effective mesh cleaning algorithm. A two-step patch cropping algorithm and a patch texture mapping module refine the size of 1-hop geodesic patches and build the relationship between the mesh geometry and color information, resulting in the generation of 1-hop textured geodesic patches. Three types of features are extracted to quantify the distortion: patch color smoothness, patch discrete mean curvature, and patch pixel color average and variance. To the best of our knowledge, GeodesicPSIM is the first model-based metric especially designed for static meshes with texture maps. GeodesicPSIM provides state-of-the-art performance in comparison with image-based, point-based, and video-based metrics on a newly created and challenging database. We also prove the robustness of GeodesicPSIM by introducing different settings of hyperparameters. Ablation studies also exhibit the effectiveness of three proposed features and the patch cropping algorithm.

Zhongxiang Lei, Qi Yang, Ping Qiu et al.

Federated optimization is a constrained form of distributed optimization that enables training a global model without directly sharing client data. Although existing algorithms can guarantee convergence in theory and often achieve stable training in practice, the reasons behind performance degradation under data heterogeneity remain unclear. To address this gap, the main contribution of this paper is to provide a theoretical perspective that explains why such degradation occurs. We introduce the assumption that heterogeneous client data lead to distinct local optima, and show that this assumption implies two key consequences: 1) the distance among clients' local optima raises the lower bound of the global objective, making perfect fitting of all client data impossible; and 2) in the final training stage, the global model oscillates within a region instead of converging to a single optimum, limiting its ability to fully fit the data. These results provide a principled explanation for performance degradation in non-iid settings, which we further validate through experiments across multiple tasks and neural network architectures. The framework used in this paper is open-sourced at: https://github.com/NPCLEI/fedtorch.

Qi Yang, Binjie Mao, Zili Wang et al.

Foley is a term commonly used in filmmaking, referring to the addition of daily sound effects to silent films or videos to enhance the auditory experience. Video-to-Audio (V2A), as a particular type of automatic foley task, presents inherent challenges related to audio-visual synchronization. These challenges encompass maintaining the content consistency between the input video and the generated audio, as well as the alignment of temporal and loudness properties within the video. To address these issues, we construct a controllable video-to-audio synthesis model, termed Draw an Audio, which supports multiple input instructions through drawn masks and loudness signals. To ensure content consistency between the synthesized audio and target video, we introduce the Mask-Attention Module (MAM), which employs masked video instruction to enable the model to focus on regions of interest. Additionally, we implement the Time-Loudness Module (TLM), which uses an auxiliary loudness signal to ensure the synthesis of sound that aligns with the video in both loudness and temporal dimensions. Furthermore, we have extended a large-scale V2A dataset, named VGGSound-Caption, by annotating caption prompts. Extensive experiments on challenging benchmarks across two large-scale V2A datasets verify Draw an Audio achieves the state-of-the-art. Project page: https://yannqi.github.io/Draw-an-Audio/.

Qi Yang, Hao Zhu

Many methods based on sparse and low-rank representation been developed along with guarantees of correct outlier detection. Self-representation states that a point in a subspace can always be expressed as a linear combination of other points in the subspace. A suitable Markov Chain can be defined on the self-representation and it allows us to recognize the difference between inliers and outliers. However, the reconstruction error of self-representation that is still informative to detect outlier detection, is neglected.Inspired by the gradient boosting, in this paper, we propose a new outlier detection framework that combines a series of weak "outlier detectors" into a single strong one in an iterative fashion by constructing multi-pass self-representation. At each stage, we construct a self-representation based on elastic-net and define a suitable Markov Chain on it to detect outliers. The residual of the self-representation is used for the next stage to learn the next weaker outlier detector. Such a stage will repeat many times. And the final decision of outliers is generated by the previous all results. Experimental results on image and speaker datasets demonstrate its superiority with respect to state-of-the-art sparse and low-rank outlier detection methods.

Zili Wang, Robert Zhang, Kun Ding et al.

Continuous visual autoregressive (AR) models have demonstrated promising performance in image generation. However, the heavy autoregressive inference burden imposes significant overhead. In Large Language Models (LLMs), speculative decoding has effectively accelerated discrete autoregressive inference. However, the absence of an analogous theory for continuous distributions precludes its use in accelerating continuous AR models. To fill this gap, this work presents continuous speculative decoding, and addresses challenges from: 1) low acceptance rate, caused by inconsistent output distribution between target and draft models, and 2) modified distribution without analytic expression, caused by complex integral. To address challenge 1), we propose denoising trajectory alignment and token pre-filling strategies. To address challenge 2), we introduce acceptance-rejection sampling algorithm with an appropriate upper bound, thereby avoiding explicitly calculating the integral. Furthermore, our denoising trajectory alignment is also reused in acceptance-rejection sampling, effectively avoiding repetitive diffusion model inference. Extensive experiments demonstrate that our proposed continuous speculative decoding achieves over $2\times$ speedup on off-the-shelf models, while maintaining the original generation quality. Codes is available at: https://github.com/MarkXCloud/CSpD

Cheng Li, Weijian Huang, Jiarun Liu et al.

The integration of medical imaging and clinical text has enabled the emergence of generalist artificial intelligence (AI) systems for healthcare. However, pervasive biases, such as imbalanced disease prevalence, skewed anatomical region distributions, heterogeneous imaging protocols, and demographic disparities, pose significant challenges to the fairness and reliability of vision-language systems in real-world clinical settings. Here we present BiasCareVL, a bias-aware multimodal learning framework that introduces bias control directly into model design, rather than treating it as a post hoc correction. BiasCareVL incorporates adaptive uncertainty modeling with optional human-in-the-loop refinement to regulate the influence of dominant data patterns and to promote equitable reasoning under distributional imbalance. Trained on 3.44 million samples spanning over 15 imaging modalities, the framework supports diverse clinical tasks, including visual question answering, disease classification, segmentation, and report generation within a unified representation space. Across eight public benchmarks covering dermatology, oncology, radiology, and pathology, BiasCareVL consistently outperforms 20 state-of-the-art methods, with pronounced gains in clinically challenging scenarios, including over 10% accuracy improvement in multi-class skin lesion diagnosis and more than 20% Dice improvement in small tumor segmentation. Furthermore, BiasCareVL achieves diagnostic performance exceeding human accuracy with substantially reduced time requirements when evaluated with board-certified radiologists. By open-sourcing BiasCareVL, we aim to promote a transparent, reproducible, and equitable future for AI in healthcare, paving the way for general-purpose, trustworthy, and clinically reliable AI systems.

Changming Xiao, Qi Yang, Feng Zhou et al.

Diffusion models have revolted the field of text-to-image generation recently. The unique way of fusing text and image information contributes to their remarkable capability of generating highly text-related images. From another perspective, these generative models imply clues about the precise correlation between words and pixels. In this work, a simple but effective method is proposed to utilize the attention mechanism in the denoising network of text-to-image diffusion models. Without re-training nor inference-time optimization, the semantic grounding of phrases can be attained directly. We evaluate our method on Pascal VOC 2012 and Microsoft COCO 2014 under weakly-supervised semantic segmentation setting and our method achieves superior performance to prior methods. In addition, the acquired word-pixel correlation is found to be generalizable for the learned text embedding of customized generation methods, requiring only a few modifications. To validate our discovery, we introduce a new practical task called "personalized referring image segmentation" with a new dataset. Experiments in various situations demonstrate the advantages of our method compared to strong baselines on this task. In summary, our work reveals a novel way to extract the rich multi-modal knowledge hidden in diffusion models for segmentation.

Mufan Liu, Qi Yang, Miaoran Zhao et al.

Implicit Neural Representations (INRs) have emerged as a powerful approach for video representation, offering versatility across tasks such as compression and inpainting. However, their implicit formulation limits both interpretability and efficacy, undermining their practicality as a comprehensive solution. We propose a novel video representation based on deformable 2D Gaussian splatting, dubbed D2GV, which aims to achieve three key objectives: 1) improved efficiency while delivering superior quality; 2) enhanced scalability and interpretability; and 3) increased friendliness for downstream tasks. Specifically, we initially divide the video sequence into fixed-length Groups of Pictures (GoP) to allow parallel training and linear scalability with video length. For each GoP, D2GV represents video frames by applying differentiable rasterization to 2D Gaussians, which are deformed from a canonical space into their corresponding timestamps. Notably, leveraging efficient CUDA-based rasterization, D2GV converges fast and decodes at speeds exceeding 400 FPS, while delivering quality that matches or surpasses state-of-the-art INRs. Moreover, we incorporate a learnable pruning and quantization strategy to streamline D2GV into a more compact representation. We demonstrate D2GV's versatility in tasks including video interpolation, inpainting and denoising, underscoring its potential as a promising solution for video representation. Code is available at: https://github.com/Evan-sudo/D2GV.

He Huang, Qi Yang, Mufan Liu et al.

Existing 4D Gaussian Splatting methods rely on per-Gaussian deformation from a canonical space to target frames, which overlooks redundancy among adjacent Gaussian primitives and results in suboptimal performance. To address this limitation, we propose Anchor-Driven Deformable and Compressed Gaussian Splatting (ADC-GS), a compact and efficient representation for dynamic scene reconstruction. Specifically, ADC-GS organizes Gaussian primitives into an anchor-based structure within the canonical space, enhanced by a temporal significance-based anchor refinement strategy. To reduce deformation redundancy, ADC-GS introduces a hierarchical coarse-to-fine pipeline that captures motions at varying granularities. Moreover, a rate-distortion optimization is adopted to achieve an optimal balance between bitrate consumption and representation fidelity. Experimental results demonstrate that ADC-GS outperforms the per-Gaussian deformation approaches in rendering speed by 300%-800% while achieving state-of-the-art storage efficiency without compromising rendering quality. The code is released at https://github.com/H-Huang774/ADC-GS.git.

Qi Yang, Le Yang, Geert Van Der Auwera et al.

Most existing 3D Gaussian Splatting (3DGS) compression schemes focus on producing compact 3DGS representation via implicit data embedding. They have long coding times and highly customized data format, making it difficult for widespread deployment. This paper presents a new 3DGS compression framework called HybridGS, which takes advantage of both compact generation and standardized point cloud data encoding. HybridGS first generates compact and explicit 3DGS data. A dual-channel sparse representation is introduced to supervise the primitive position and feature bit depth. It then utilizes a canonical point cloud encoder to perform further data compression and form standard output bitstreams. A simple and effective rate control scheme is proposed to pivot the interpretable data compression scheme. At the current stage, HybridGS does not include any modules aimed at improving 3DGS quality during generation. But experiment results show that it still provides comparable reconstruction performance against state-of-the-art methods, with evidently higher encoding and decoding speed. The code is publicly available at https://github.com/Qi-Yangsjtu/HybridGS.

Wenjie Huang, Qi Yang, Shuting Xia et al.

Learning-based point cloud compression presents superior performance to handcrafted codecs. However, pretrained-based methods, which are based on end-to-end training and expected to generalize to all the potential samples, suffer from training data dependency. Implicit neural representation (INR) based methods are distribution-agnostic and more robust, but they require time-consuming online training and suffer from the bitstream overhead from the overfitted model. To address these limitations, we propose HybridINR-PCGC, a novel hybrid framework that bridges the pretrained model and INR. Our framework retains distribution-agnostic properties while leveraging a pretrained network to accelerate convergence and reduce model overhead, which consists of two parts: the Pretrained Prior Network (PPN) and the Distribution Agnostic Refiner (DAR). We leverage the PPN, designed for fast inference and stable performance, to generate a robust prior for accelerating the DAR's convergence. The DAR is decomposed into a base layer and an enhancement layer, and only the enhancement layer needed to be packed into the bitstream. Finally, we propose a supervised model compression module to further supervise and minimize the bitrate of the enhancement layer parameters. Based on experiment results, HybridINR-PCGC achieves a significantly improved compression rate and encoding efficiency. Specifically, our method achieves a Bpp reduction of approximately 20.43% compared to G-PCC on 8iVFB. In the challenging out-of-distribution scenario Cat1B, our method achieves a Bpp reduction of approximately 57.85% compared to UniPCGC. And our method exhibits a superior time-rate trade-off, achieving an average Bpp reduction of 15.193% relative to the LINR-PCGC on 8iVFB.