SCALAR++: Efficient Controllable Generation via Scale-wise Visual Autoregressive Learning

Ryan Xu Dong-Yang Jin Shawn Chen Yancheng Bai Jingzhe Ma Rui Lan Jianhao Zeng Yun-Yang Ge Mingzhe Zheng Lei Sun Xiangxiang Chu

Aug 2026 · International Journal of Computer Vision · Vol 134 · 1 citation · 54 references

TL;DR

This work proposes a Scale-wise Conditional Decoding mechanism, which projects semantic signals from a frozen vision encoder into scale-specific layers of the VAR backbone, and introduces a Unified Control Alignment strategy (SCALAR-Uni) to handle diverse control modalities within a single projection space.

DOI Semantic Scholar

View source

Similar papers

Jul 2026

UniGen-AR: Unifying Visual Generation with Auto-Regressive Modeling

Unigen-AR, a framework that pairs a general-purpose multi-modal language model (MLLM) with an efficient next-scale visual auto-regressive (VAR) decoder, is proposed, establishing visual auto-regressive modeling as a compelling and efficient backbone for unified visual generation.

Zhipeng Bao, Zhen Zhu, Nupur Kumari et al. · arXiv.org · 0 citations

2025

FreqExit: Enabling Early-Exit Inference for Visual Autoregressive Models via Frequency-Aware Guidance

This work proposes FreqExit, an integrated loss design that enables dynamic inference in VAR without altering its architecture or compromising output quality, and designs targeted mechanisms that guide the model to learn more effectively through frequency-aware supervision.

Ying Li, Chengfei Lyu, Huan Wang · Neural Information Processin... · 6 citations · ⚡1

Aug 2026

4DWeaver: Bridging Reconstruction and Generation Via Compact Autoregressive Priors.

This work proposes Compact Autoregressive Latent Prior (CALP), which regularizes low-dimensional latent variables with a history-conditioned autoregressive prior, achieving compact, long-range temporally coherent, and spatially structured latent representations through a more reasonable latent capacity allocation and a...

Ruosen Hao, Tian-Yu Shen, Jing-Yuan Gao et al. · IEEE Transactions on Pattern... · 0 citations

Preprint Aug 2026

VISTA: Test-Time Compositional Alignment for Visual Autoregressive Generation

VISTA is the first gradient-based test-time alignment framework for next-scale autoregressive image generation, and introduces the mechanisms needed to make such optimization stable across scales, together with an extensible objective space that any differentiable constraint on cross-attention can plug into.

Hossein Shahabadi, Niki Sepasian, Mahdieh Soleymani Baghshah · 0 citations

Jul 2026

Next-Dense-Stride Prediction for Multimodal Autoregressive Visual Modeling

DenseAR is extended to a unified model that handles multiple modalities and imaging tasks within a single backbone that unifies cross-modal translation, modality-conditioned generation, and tumor segmentation, while remaining competitive with task-specific methods.

Chicago Y. Park, Jia-Lin Mao, Xiaojian Xu et al. · arXiv.org · 0 citations

Preprint Aug 2026

MotionCraft: Latent World Modeling with Sparse Attention for Visual Upscaling

MotionCraft is presented, a controllable VSR framework that formulates restoration as motion-aware latent state prediction inspired by world models and integrates adaptive sparse attention with an explicit user-accessible control interface to deliver temporally consistent, high-quality reconstructions under streaming c...

Rong Fu, Chun-Lei Meng, Yangcheng Zeng et al. · 1 citation