Beyond Tokens: A Survey on Decoding Methods for Large Language and Vision-Language Models
ACM SIGKDD Explorations Newsletter · 17.06.2026 · 28 (1) · 1-20
DOI: 10.1145/3820356.3820357 ↗Kurz gefasst
Fachartikel in der Zeitschrift ACM SIGKDD Explorations Newsletter, online erschienen am 17.06.2026. Verfasst von Haoran Wang und 3 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 188 Literaturangaben.
Abstract (Originalsprache)
Large language models (LLMs) and large vision-language models (LVLMs) have demonstrated impressive generative capabilities, yet ensuring their outputs align with user intent is still challenging. While most existing approaches address this issue at the training stage, inference-time approaches like decoding methods offer a more efficient and scalable solution. Decoding methods control model generation by guiding token-level selection, performing sequencelevel generation, or generating tokens in parallel to accelerate the process. In this survey, we identify three emerging paradigms from recent works on decoding methods for LLMs and LVLMs, provide a systematic review of these methods, highlight ongoing challenges, and discuss potential future research directions. Our goal is to underscore the efficiency and effectiveness of decoding methods and offer a practical view of their applications. Paper lists and more resources on decoding methods for LLMs and LVLMs can be found at https://github.com/wang2226/Awesome-LLM-Decoding.
Zitieren
APA
Wang, H., Xu, X., Yu, P. S., & Shu, K. (2026). Beyond Tokens: A Survey on Decoding Methods for Large Language and Vision-Language Models. ACM SIGKDD Explorations Newsletter, 28(1), 1-20. https://doi.org/10.1145/3820356.3820357
Vancouver
Wang H, Xu X, Yu PS, Shu K. Beyond Tokens: A Survey on Decoding Methods for Large Language and Vision-Language Models. ACM SIGKDD Explorations Newsletter. 2026;28(1):1-20. doi:10.1145/3820356.3820357
BibTeX
@article{Wang2026,
author = {Wang, Haoran and Xu, Xiongxiao and Yu, Philip S. and Shu, Kai},
title = {Beyond Tokens: A Survey on Decoding Methods for Large Language and Vision-Language Models},
journal = {ACM SIGKDD Explorations Newsletter},
year = {2026},
volume = {28},
number = {1},
pages = {1--20},
doi = {10.1145/3820356.3820357}
}
EUVIMED