Keyu Fan

Zunhai Su, Hengyuan Zhang, Wei Wu et al.

As the foundational architecture of modern machine learning, Transformers have driven remarkable progress across diverse AI domains. Despite their transformative impact, a persistent challenge across various Transformers is Attention Sink (AS), in which a disproportionate amount of attention is focused on a small subset of specific yet uninformative tokens. AS complicates interpretability, significantly affecting the training and inference dynamics, and exacerbates issues such as hallucinations. In recent years, substantial research has been dedicated to understanding and harnessing AS. However, a comprehensive survey that systematically consolidates AS-related research and offers guidance for future advancements remains lacking. To address this gap, we present the first survey on AS, structured around three key dimensions that define the current research landscape: Fundamental Utilization, Mechanistic Interpretation, and Strategic Mitigation. Our work provides a pivotal contribution by clarifying key concepts and guiding researchers through the evolution and trends of the field. We envision this survey as a definitive resource, empowering researchers and practitioners to effectively manage AS within the current Transformer paradigm, while simultaneously inspiring innovative advancements for the next generation of Transformers. The paper list of this work is available at https://github.com/ZunhaiSu/Awesome-Attention-Sink.

Zunhai Su, Weihao Ye, Hansen Feng et al.

Learning-based 3D visual geometry models have significantly advanced with the advent of large-scale transformers. Among these, StreamVGGT leverages frame-wise causal attention to deliver robust and efficient streaming 3D reconstruction. However, it suffers from unbounded growth in the Key-Value (KV) cache due to the massive influx of vision tokens from multi-image and long-video inputs, leading to increased memory consumption and inference latency as input frames accumulate. This ultimately limits its scalability for long-horizon applications. To address this gap, we propose XStreamVGGT, a tuning-free approach that seamlessly integrates pruning and quantization to systematically compress the KV cache, enabling extremely memory-efficient streaming inference. Specifically, redundant KVs generated from multi-frame inputs are initially pruned to conform to a fixed KV memory budget using an efficient token-importance identification mechanism that maintains full compatibility with high-performance attention kernels (e.g., FlashAttention). Additionally, leveraging the inherent distribution patterns of KV tensors, we apply dimension-adaptive KV quantization within the pruning pipeline to further minimize memory overhead while preserving numerical accuracy. Extensive evaluations show that XStreamVGGT achieves mostly negligible performance degradation while substantially reducing memory usage by 4.42$\times$ and accelerating inference by 5.48$\times$, enabling practical and scalable streaming 3D applications. The code is available at https://github.com/ywh187/XStreamVGGT/.

Yifan Liu, Keyu Fan, Weihao Yu et al.

Recent advances in generalizable 3D Gaussian Splatting have demonstrated promising results in real-time high-fidelity rendering without per-scene optimization, yet existing approaches still struggle to handle unfamiliar visual content during inference on novel scenes due to limited generalizability. To address this challenge, we introduce MonoSplat, a novel framework that leverages rich visual priors from pre-trained monocular depth foundation models for robust Gaussian reconstruction. Our approach consists of two key components: a Mono-Multi Feature Adapter that transforms monocular features into multi-view representations, coupled with an Integrated Gaussian Prediction module that effectively fuses both feature types for precise Gaussian generation. Through the Adapter's lightweight attention mechanism, features are seamlessly aligned and aggregated across views while preserving valuable monocular priors, enabling the Prediction module to generate Gaussian primitives with accurate geometry and appearance. Through extensive experiments on diverse real-world datasets, we convincingly demonstrate that MonoSplat achieves superior reconstruction quality and generalization capability compared to existing methods while maintaining computational efficiency with minimal trainable parameters. Codes are available at https://github.com/CUHK-AIM-Group/MonoSplat.

Yi Liu, Haoxuan Xu, Hongbo Duan et al.

Visual SLAM algorithms achieve significant improvements through the exploration of 3D Gaussian Splatting (3DGS) representations, particularly in generating high-fidelity dense maps. However, they depend on a static environment assumption and experience significant performance degradation in dynamic environments. This paper presents GGD-SLAM, a framework that employs a generalizable motion model to address the challenges of localization and dense mapping in dynamic environments - without predefined semantic annotations or depth input. Specifically, the proposed system employs a First-In-First-Out (FIFO) queue to manage incoming frames, facilitating dynamic semantic feature extraction through a sequential attention mechanism. This is integrated with a dynamic feature enhancer to separate static and dynamic components. Additionally, to minimize dynamic distractors' impact on the static components, we devise a method to fill occluded areas via static information sampling and design a distractor-adaptive Structure Similarity Index Measure (SSIM) loss tailored for dynamic environments, significantly enhancing the system's resilience. Experiments conducted on real-world dynamic datasets demonstrate that the proposed system achieves state-of-the-art performance in camera pose estimation and dense reconstruction in dynamic scenes.

Yifu Luo, Xinhao Hu, Keyu Fan et al.

Reinforcement learning (RL) has garnered increasing attention in text-to-image (T2I) generation. However, most existing RL approaches are tailored to either diffusion models or autoregressive models, overlooking an important alternative: masked generative models. In this work, we propose Mask-GRPO, the first method to incorporate Group Relative Policy Optimization (GRPO)-based RL into this overlooked paradigm. Our core insight is to redefine the transition probability, which is different from current approaches, and formulate the unmasking process as a multi-step decision-making problem. To further enhance our method, we explore several useful strategies, including removing the KL constraint, applying the reduction strategy, and filtering out low-quality samples. Using Mask-GRPO, we improve a base model, Show-o, with substantial improvements on standard T2I benchmarks and preference alignment, outperforming existing state-of-the-art approaches. The code is available on https://github.com/xingzhejun/Mask-GRPO

Yingshan Liang, Keyu Fan, Zhicheng Du et al.

Video-to-audio (V2A) generation shows great potential in fields such as film production. Despite significant advances, current V2A methods relying on global video information struggle with complex scenes and generating audio tailored to specific objects. To address these limitations, we introduce Hear-Your-Click, an interactive V2A framework enabling users to generate sounds for specific objects by clicking on the frame. To achieve this, we propose Object-aware Contrastive Audio-Visual Fine-tuning (OCAV) with a Mask-guided Visual Encoder (MVE) to obtain object-level visual features aligned with audio. Furthermore, we tailor two data augmentation strategies, Random Video Stitching (RVS) and Mask-guided Loudness Modulation (MLM), to enhance the model's sensitivity to segmented objects. To measure audio-visual correspondence, we designed a new evaluation metric, the CAV score. Extensive experiments demonstrate that our framework offers more precise control and improves generation performance across various metrics. Project Page: https://github.com/SynapGrid/Hear-Your-Click

Yuxin Zhang, Ziyu Lu, Hongbo Duan et al.

Despite recent advancements in neural 3D reconstruction, the dependence on dense multi-view captures restricts their broader applicability. Additionally, 3D scene generation is vital for advancing embodied AI and world models, which depend on diverse, high-quality scenes for learning and evaluation. In this work, we propose Gen3d, a novel method for generation of high-quality, wide-scope, and generic 3D scenes from a single image. After the initial point cloud is created by lifting the RGBD image, Gen3d maintains and expands its world model. The 3D scene is finalized through optimizing a Gaussian splatting representation. Extensive experiments on diverse datasets demonstrate the strong generalization capability and superior performance of our method in generating a world model and Synthesizing high-fidelity and consistent novel views.