LANTERN++: Enhancing Relaxed Speculative Decoding with Static Tree Drafting for Visual Auto-regressive Models

Sihwan Park, Doohyuk Jang, Sung-Yub Kim, Souvik Kundu, Eunho Yang

Track: Main paper track (up to 5 pages excluding references and appendix)

Keywords: Speculative Decoding, Multi-modal Auto-regressive Models

Abstract: Speculative decoding has been widely used to accelerate autoregressive (AR) text generation. However, its effectiveness for visual AR models remains limited due to token selection ambiguity, where multiple tokens share similarly low probabilities and thus reduce acceptance rates. Recently, relaxed speculative decoding with dynamic tree drafting was proposed to mitigate this ambiguity, demonstrating promising results in accelerating visual AR models. Nevertheless, we observe that token selection ambiguity continues to negatively affect dynamic tree drafting, resulting in shallow draft trees and limited acceleration. To overcome this issue, we introduce LANTERN++, a novel framework that integrates static tree drafting with a relaxed acceptance condition, allowing drafts to be selected independently of low-confidence predictions. This enables the acceptance of deeper sequences, improving decoding efficiency while preserving image quality. Extensive experiments on state-of-the-art visual AR models demonstrate that LANTERN++ significantly accelerates inference, achieving up to $\mathbf{\times 2.56}$ speedup over standard AR decoding while maintaining high image quality. The code is publicly available at \url{https://github.com/jadohu/LANTERN}.

Anonymization: This submission has been anonymized for double-blind review via the removal of identifying information such as names, affiliations, and identifying URLs.

Submission Number: 85