ExtraVAR: Stage-Aware RoPE Remapping for Resolution Extrapolation in Visual Autoregressive Models

TL;DR

ExtraVAR introduces a stage-aware RoPE remapping and adaptive attention calibration to improve high-resolution image synthesis in visual autoregressive models, addressing failure modes caused by scale-wise band mismatches.

Contribution

The paper proposes a novel, training-free method for resolution extrapolation in VAR models that suppresses failure modes by remapping frequency bands and calibrates attention dispersion adaptively.

Findings

Outperforms prior methods in structural coherence.

Enhances fine-detail fidelity at higher resolutions.

Effectively suppresses repetition and detail degradation.

Abstract

Visual Autoregressive (VAR) models have emerged as a strong alternative to diffusion for image synthesis, yet their fixed training resolution prevents direct generation at higher resolutions. Naively transferring training-free extrapolation methods from LLMs or diffusion models to VAR yields three characteristic failure modes: global repetition, local repetition, and detail degradation. We trace them to a unified band-stage mismatch: VAR generates images in a coarse-to-fine, scale-wise process where each stage is driven by a distinct dominant RoPE frequency band, and each failure mode emerges when the dominant band of a particular stage is disrupted. Building on this insight, we propose Stage-Aware RoPE Remapping, a training-free strategy that assigns each frequency band a stage-specific remapping rule, jointly suppressing all three failure modes. We further observe that attention becomes systematically dispersed as the image resolution increases. Existing methods typically depend on predefined attention scaling factors, which are neither adaptive to the target resolution nor capable of faithfully capturing the actual extent of attention dispersion. We therefore propose Entropy-Driven Adaptive Attention Calibration, which quantifies dispersion via a resolution-invariant normalized entropy and yields a closed-form per-head scaling factor that realigns the extrapolated-resolution attention entropy with its training-resolution counterpart. Extensive experiments show that our method consistently outperforms prior resolution-extrapolation methods in both structural coherence and fine-detail fidelity. Our code is available at https://github.com/feihongyan1/ExtraVAR.