paper-with-me

홈 › Papers

Direction-Aware Diagonal Autoregressive Image Generation

2025-03-14 · Yijia Xu, Jianzhong Ju, Jian Luan, Jinshi Cui

The raster-ordered image token sequence exhibits a significant Euclidean distance between index-adjacent tokens at line breaks, making it unsuitable for autoregressive generation. To address this issue, this paper proposes Direction-Aware Diagonal Autoregressive Image Generation (DAR) method, which generates image tokens following a diagonal scanning order. The proposed diagonal scanning order ensures that tokens with adjacent indices remain in close proximity while enabling causal attention to gather information from a broader range of directions. Additionally, two direction-aware modules: 4D-RoPE and direction embeddings are introduced, enhancing the model's capability to handle frequent changes in generation direction. To leverage the representational capacity of the image tokenizer, we use its codebook as the image token embeddings. We propose models of varying scales, ranging from 485M to 2.0B. On the 256$\times$256 ImageNet benchmark, our DAR-XL (2.0B) outperforms all previous autoregressive image generators, achieving a state-of-the-art FID score of 1.37.

📄 PDF Abstract BibTeX arXiv:2503.11129

Code (0)

등록된 구현이 없습니다.

Tasks

Image Generation

Methods 이 논문이 사용한 방법론

Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Attention 설명 없음

Similar Papers 제목 키워드 기반

Fast Autoregressive Video Generation with Diagonal Decoding

2025-03-18 · Yang Ye, Junliang Guo, HaoYu Wu, Tianyu He 외

Autoregressive Transformer models have demonstrated impressive performance in video generation, but their sequential token-by-token decoding process poses a major bottleneck, particularly for long videos represented by t…

Video Generation

NUWA-Infinity: Autoregressive over Autoregressive Generation for Infinite Visual Synthesis

2022-07-20 · Chenfei Wu, Jian Liang, Xiaowei Hu, Zhe Gan 외

In this paper, we present NUWA-Infinity, a generative model for infinite visual synthesis, which is defined as the task of generating arbitrarily-sized high-resolution images or long-duration videos. An autoregressive ov…

Image OutpaintingText-to-Image GenerationVideo Generation

Head-Aware KV Cache Compression for Efficient Visual Autoregressive Modeling

2025-04-12 · Ziran Qin, Youru Lv, Mingbao Lin, Zeren Zhang 외

Visual Autoregressive (VAR) models have emerged as a powerful approach for multi-modal content creation, offering high efficiency and quality across diverse multimedia applications. However, they face significant memory …

Bidirectional Representations Augmented Autoregressive Biological Sequence Generation

2025-10-09 · Xiang Zhang, Jiaqi Wei, Zijie Qiu, Sheng Xu 외 arxiv

Autoregressive (AR) models, common in sequence generation, are limited in many biological tasks such as de novo peptide sequencing and protein modeling by their unidirectional nature, failing to capture crucial global bi…

Autoregressive Omni-Aware Outpainting for Open-Vocabulary 360-Degree Image Generation

2023-09-07 · Zhuqiang Lu, Kun Hu, Chaoyue Wang, Lei Bai 외

A 360-degree (omni-directional) image provides an all-encompassing spherical view of a scene. Recently, there has been an increasing interest in synthesising 360-degree images from conventional narrow field of view (NFoV…

Image Generation