Not All Tokens are Guided Equal: Elucidating Guidance in Autoregressive Modelling

ICLR 2026 Conference Submission12747 Authors

Keywords: diffusion models, autoregressive models, guidance

TL;DR: This paper sheds light on a key issue with guidance in AR modelling and proposes a new guidance method to bring the performance of AR models closer to diffusion models.

Abstract: Autoregressive (AR) models based on next-scale prediction are rapidly emerging as a powerful tool for image generation, but they face a critical weakness: information inconsistencies between patches across timesteps introduced by progressive resolution scaling. These inconsistencies scatter guidance signals, causing them to drift away from conditioning information and leaving behind ambiguous, unfaithful features. We tackle this challenge with Information-Grounding Guidance (IGG) — a novel mechanism that anchors guidance to semantically important regions through attention. By adaptively reinforcing informative patches during sampling, IGG ensures that guidance and content remain tightly aligned. Across both class-conditioned and text-to-image generation tasks, IGG delivers sharper, more coherent, and semantically grounded images, setting a new benchmark for AR-based methods.

Primary Area: generative models

Submission Number: 12747