Autoregressive models have recently shown great promise in visual generation by leveraging discrete token sequences akin to language modeling. However, existing approaches typically rely on fixed r...
<p><strong>Tác giả:</strong> Ruiqing Yang, Kaixin Zhang, Zheng Zhang, Shan You, Tao Huang</p>
<p>Autoregressive models have recently shown great promise in visual generation by leveraging discrete token sequences akin to language modeling. However, existing approaches typically rely on fixed raster-scan orderings and token-by-token decoding or the complexity of multi-scale representations, which impose suboptimal factorizations of the image distribution and limit both efficiency and scalability. In this work, we propose Expanding Autoregressive Representation (EAR), a novel autoregressiv...</p>
<p>📄 DOI: <a href="https://doi.org/https://doi.org/10.1145/3767308.3835202" target="_blank" rel="noopener">https://doi.org/10.1145/3767308.3835202</a></p>
<p>📖 <a href="https://doi.org/https://doi.org/10.1145/3767308.3835202" target="_blank" rel="noopener">Xem bài báo đầy đủ</a></p>
Bài viết được biên soạn lại bởi LIFVietnam. Nguồn tham khảo: OpenAlex
