Related papers: ExtraVAR: Stage-Aware RoPE Remapping for Resolutio…

Resolution generalization in image generation tasks enables the production of higher-resolution images with lower training resolution overhead. However, a key obstacle for diffusion transformers in addressing this problem is the mismatch…

Current autoregressive video diffusion models are constrained by three core bottlenecks: (i) the finite temporal horizon imposed by the base model's 3D Rotary Positional Embedding (3D-RoPE), (ii) slow prompt responsiveness in maintaining…

Generative models for Image Super-Resolution (SR) are increasingly powerful, yet their reliance on self-attention's quadratic complexity (O(N^2)) creates a major computational bottleneck. Linear Attention offers an O(N) solution, but its…

Diffusion models have recently achieved remarkable success in generative modeling, yet their training dynamics across different noise levels remain highly imbalanced, which can lead to inefficient optimization and unstable learning…

Vision-language models (VLMs) exhibit remarkable zero-shot capabilities but struggle with distribution shifts in downstream tasks when labeled data is unavailable, which has motivated the development of Test-Time Adaptation (TTA) to improve…

Large multimodal models (LMMs) "see" images by leveraging the attention mechanism between text and visual tokens in the transformer decoder. Ideally, these models should focus on key visual information relevant to the text token. However,…

Multi-view deep neural network is perhaps the most successful approach in 3D shape classification. However, the fusion of multi-view features based on max or average pooling lacks a view selection mechanism, limiting its application in,…

This work presents the network architecture EVP (Enhanced Visual Perception). EVP builds on the previous work VPD which paved the way to use the Stable Diffusion network for computer vision tasks. We propose two major enhancements. First,…

Audio-visual emotion recognition (AVER) methods typically fuse utterance-level features, and even frame-level attention models seldom address the frame-rate mismatch across modalities. In this paper, we propose a Transformer-based framework…

Autoregressive (AR)-Diffusion hybrid paradigms combine AR's structured semantic modeling with diffusion's high-fidelity synthesis, yet suffer from a dual speed bottleneck: the sequential AR stage and the iterative multi-step denoising of…

We prove under practical assumptions that Rotary Positional Embedding (RoPE) introduces an intrinsic distance-dependent bias in attention scores that limits RoPE's ability to model long-context. RoPE extension methods may alleviate this…

We improve upon the two-stage sparse vector autoregression (sVAR) method in Davis et al. (2016) by proposing an alternative two-stage modified sVAR method which relies on time series graphical lasso to estimate sparse inverse spectral…

Group-based policy optimization methods like GRPO and GSPO have become standard for training multimodal models, leveraging group-wise rollouts and relative advantage estimation. However, they suffer from a critical \emph{gradient vanishing}…

Generative models trained on extensive high-quality datasets effectively capture the structural and statistical properties of clean images, rendering them powerful priors for transforming degraded features into clean ones in image…

Traditional approaches to interpolate/extrapolate frames in a video sequence require accurate pixel correspondences between images, e.g., using optical flow. Their results stem on the accuracy of optical flow estimation, and could generate…

Conventional wisdom suggests that autoregressive models are used to process discrete data. When applied to continuous modalities such as visual data, Visual AutoRegressive modeling (VAR) typically resorts to quantization-based approaches to…

Auto-regressive (AR) models, initially successful in language generation, have recently shown promise in visual generation tasks due to their superior sampling efficiency. Unlike image generation, video generation requires a substantially…

In this paper, we investigate the continual learning of Vision Transformers (ViT) for the challenging exemplar-free scenario, with special focus on how to efficiently distill the knowledge of its crucial self-attention mechanism (SAM). Our…

Learning-based multi-view stereo (MVS) method heavily relies on feature matching, which requires distinctive and descriptive representations. An effective solution is to apply non-local feature aggregation, e.g., Transformer. Albeit useful,…

Large vision-language models (LVLMs) are typically trained using autoregressive language modeling objectives, which align visual representations with linguistic space. While effective for multimodal reasoning, this alignment can weaken…