Adaptive Visual Autoregressive Acceleration via Dual-Linkage Entropy Analysis

TL;DR

NOVA is a training-free, entropy-based token reduction framework that adaptively accelerates visual autoregressive models during inference by pruning low-entropy tokens and dynamically adjusting reduction ratios.

Contribution

It introduces NOVA, a novel entropy analysis method for adaptive, training-free token reduction in VAR models, overcoming limitations of heuristic and non-adaptive approaches.

Findings

Significant inference acceleration demonstrated across multiple experiments.

Maintains high generation quality despite token pruning.

Effectively identifies optimal acceleration points via entropy inflection.

Abstract

Visual AutoRegressive modeling (VAR) suffers from substantial computational cost due to the massive token count involved. Failing to account for the continuous evolution of modeling dynamics, existing VAR token reduction methods face three key limitations: heuristic stage partition, non-adaptive schedules, and limited acceleration scope, thereby leaving significant acceleration potential untapped. Since entropy variation intrinsically reflects the transition of predictive uncertainty, it offers a principled measure to capture modeling dynamics evolution. Therefore, we propose NOVA, a training-free token reduction acceleration framework for VAR models via entropy analysis. NOVA adaptively determines the acceleration activation scale during inference by online identifying the inflection point of scale entropy growth. Through scale-linkage and layer-linkage ratio adjustment, NOVA dynamically computes distinct token reduction ratios for each scale and layer, pruning low-entropy tokens while reusing the cache derived from the residuals at the prior scale to accelerate inference and maintain generation quality. Extensive experiments and analyses validate NOVA as a simple yet effective training-free acceleration framework.