Xiaobin Hu

Memory in the Age of AI Agents

Yuyang Hu, Shichun Liu, Yanwei Yue, Guibin Zhang et autres

Memory has emerged, and will continue to remain, a core capability of foundation model-based agents. As research on agent memory rapidly expands and attracts unprecedented attention, the field has also become increasingly fragmented. Existing works that fall under the umbrella of agent …

Med-CMR: A Fine-Grained Benchmark Integrating Visual Evidence and Clinical Logic for Medical Complex Multimodal Reasoning

Hengfeng Gong, Xiaozhong Ji, Yuansen Liu, Wenbin Wu et autres

MLLMs MLLMs are beginning to appear in clinical workflows, but their ability to perform complex medical reasoning remains unclear. We present Med-CMR, a fine-grained Medical Complex Multimodal Reasoning benchmark. Med-CMR distinguishes from existing counterparts by three core features: 1) Systematic capability decomposition, …

Towards One-step Causal Video Generation via Adversarial Self-Distillation

Huayang Huang, Peng Xu, Xiaobin Hu, Donghao Luo et autres

Recent hybrid video generation models combine autoregressive temporal dynamics with diffusion-based spatial denoising, but their sequential, iterative nature leads to error accumulation and long inference times. In this work, we propose a distillation-based framework for efficient causal video generation that enables high-quality …

Disentangle Identity, Cooperate Emotion: Correlation-Aware Emotional Talking Portrait Generation

Weipeng Tan, Chuming Lin, Chengming Xu, FeiFan Xu et autres

Recent advances in Talking Head Generation (THG) have achieved impressive lip synchronization and visual quality through diffusion models; yet existing methods struggle to generate emotionally expressive portraits while preserving speaker identity. We identify three critical limitations in current emotional talking head generation: …

cn (code pays fourni par la source)

Identity-Preserving Text-to-Video Generation Guided by Simple yet Effective Spatial-Temporal Decoupled Representations

Yuji Wang, Moran Li, Xiaobin Hu, Ran Yi et autres

Identity-preserving text-to-video (IPT2V) generation, which aims to create high-fidelity videos with consistent human identity, has become crucial for downstream applications. However, current end-to-end frameworks suffer a critical spatial-temporal trade-off: optimizing for spatially coherent layouts of key elements ( e.g., character identity preservation) …

cn (code pays fourni par la source)

ILVMamba: Illumination-Aware Lightweight Visual Mamba Framework for Efficient High-Resolution Image Enhancement

Mingyu Liu, Jiong Xu, Yuning Cui, Xiaobin Hu et autres

Real-world image quality is often degraded by suboptimal lighting conditions, such as low light and vignetting. With advancements in sensor technology, there is an increasing demand for efficient algorithms capable of processing high-resolution images. However, existing approaches primarily focus on low-resolution data …

in, de, cn (code pays fourni par la source)

DVHGNN: Multi-Scale Dilated Vision HGNN for Efficient Vision Recognition

Caoshuo Li, Tanzhe Li, Xiaobin Hu, Donghao Luo et autres

Recently, Vision Graph Neural Network (ViG) has gained considerable attention in computer vision. Despite its groundbreaking innovation, Vision Graph Neural Network encounters key issues including the quadratic computational complexity caused by its K-Nearest Neighbor (KNN) graph construction and the limitation of pairwise …

cn (code pays fourni par la source)

Anomaly Detection in Medical Images Using Encoder-Attention-2Decoders Reconstruction

Peng Tang, Xiaoxiao Yan, Xiaobin Hu, Kai Wu et autres

Anomaly detection (AD) in medical applications is a promising field, offering a cost-effective alternative to labor-intensive abnormal data collection and labeling. However, the success of feature reconstruction-based methods in AD is often hindered by two critical factors: the domain gap of pre-trained …

de, cn, ch (code pays fourni par la source)

Disentangle Identity, Cooperate Emotion: Correlation-Aware Emotional Talking Portrait Generation

Weipeng Tan, Chuming Lin, Chengming Xu, FeiFan Xu et autres

Recent advances in Talking Head Generation (THG) have achieved impressive lip synchronization and visual quality through diffusion models; yet existing methods struggle to generate emotionally expressive portraits while preserving speaker identity. We identify three critical limitations in current emotional talking head generation: …

DVHGNN: Multi-Scale Dilated Vision HGNN for Efficient Vision Recognition

Caoshuo Li, Tanzhe Li, Xiaobin Hu, Donghao Luo et autres

Recently, Vision Graph Neural Network (ViG) has gained considerable attention in computer vision. Despite its groundbreaking innovation, Vision Graph Neural Network encounters key issues including the quadratic computational complexity caused by its K-Nearest Neighbor (KNN) graph construction and the limitation of pairwise …

3D Priors-Guided Diffusion for Blind Face Restoration

Xiaobin Lu, Xiaobin Hu, Jun Luo, Yaping Ruan et autres

Blind face restoration endeavors to restore a clear face image from a degraded counterpart. Recent approaches employing Generative Adversarial Networks (GANs) as priors have demonstrated remarkable success in this field. However, these methods encounter challenges in achieving a balance between realism and …

cn (code pays fourni par la source)

3D Priors-Guided Diffusion for Blind Face Restoration

Xiaobin Lu, Xiaobin Hu, Jun Luo, Ben Zhu et autres

Blind face restoration endeavors to restore a clear face image from a degraded counterpart. Recent approaches employing Generative Adversarial Networks (GANs) as priors have demonstrated remarkable success in this field. However, these methods encounter challenges in achieving a balance between realism and …