OPT-Tree: Speculative Decoding with Adaptive Draft Tree Structure
Transactions of the Association for Computational Linguistics · 28.02.2025 · 13 · 188-199
DOI: 10.1162/tacl_a_00735 ↗Kurz gefasst
Fachartikel in der Zeitschrift Transactions of the Association for Computational Linguistics, online erschienen am 28.02.2025. Verfasst von Jikai Wang und 7 weiteren Autor:innen. Die Arbeit wurde bisher 14-mal zitiert (Crossref) und nennt 23 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Autoregressive language models demonstrate excellent performance in various scenarios. However, the inference efficiency is limited by its one-step-one-word generation mode, which has become a pressing problem recently as the models become increasingly larger. Speculative decoding employs a “draft and then verify” mechanism to allow multiple tokens to be generated in one step, realizing lossless acceleration. Existing methods mainly adopt fixed heuristic draft structures, which do not adapt to different situations to maximize the acceptance length during verification. To alleviate this dilemma, we propose OPT-Tree, an algorithm to construct adaptive and scalable draft trees, which can be applied to any autoregressive draft model. It searches the optimal tree structure that maximizes the mathematical expectation of the acceptance length in each decoding step. Experimental results reveal that OPT-Tree outperforms the existing draft structures and achieves a speed-up ratio of up to 3.2 compared with autoregressive decoding. If the draft model is powerful enough and the node budget is sufficient, it can generate more than ten tokens in a single step. Our code is available at https://github.com/Jikai0Wang/OPT-Tree.
Zitieren
APA
Wang, J., Su, Y., Li, J., Xia, Q., Ye, Z., Duan, X., Wang, Z., & Zhang, M. (2025). OPT-Tree: Speculative Decoding with Adaptive Draft Tree Structure. Transactions of the Association for Computational Linguistics, 13, 188-199. https://doi.org/10.1162/tacl_a_00735
Vancouver
Wang J, Su Y, Li J, Xia Q, Ye Z, Duan X, et al. OPT-Tree: Speculative Decoding with Adaptive Draft Tree Structure. Transactions of the Association for Computational Linguistics. 2025;13:188-199. doi:10.1162/tacl_a_00735
BibTeX
@article{Wang2025,
author = {Wang, Jikai and Su, Yi and Li, Juntao and Xia, Qingrong and Ye, Zi and Duan, Xinyu and Wang, Zhefeng and Zhang, Min},
title = {OPT-Tree: Speculative Decoding with Adaptive Draft Tree Structure},
journal = {Transactions of the Association for Computational Linguistics},
year = {2025},
volume = {13},
pages = {188--199},
doi = {10.1162/tacl_a_00735}
}
EUVIMED