Domino: Decoupling Causal Modeling from Autoregressive Drafting in Speculative Decoding
Jianuo Huang 1,2* Yaojie Zhang 1,3* Qituan Zhang 4 Hao Lin 2 Hanlin Xu 5 Linfeng Zhang 1† 1 EPIC Lab, Shanghai Jiao Tong University 2 School of Software Engineering, HUST 3 UESTC 4 Fudan University 5 Huawei * Equal contribution. † Corresponding author · 2026
Read Domino: Decoupling Causal Modeling from Autoregressive Drafting in Speculative Decoding by Jianuo Huang 1,2* Yaojie Zhang 1,3* Qituan Zhang 4 Hao Lin 2 Hanlin Xu 5 Linfeng Zhang 1† 1 EPIC Lab, Shanghai Jiao Tong University 2 School of Software Engineering, HUST 3 UESTC 4 Fudan University 5 Huawei * Equal contribution. † Corresponding author (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-29707.txt.
Description
Jianuo Huang and five colleagues split two jobs that speculative decoding usually tangles: mapping causal links among draft tokens and running the draft itself. Their system, Domino, first drafts a whole block in parallel, then polishes it with a light head that adds prefix-based causal detail. A staged training plan first firms up the parallel base, then shifts weight toward the causally fixed final output. On Qwen3 models the scheme ran over five times faster end to end under one backend and near six times the throughput under another.
Catalog record
| CREATOR | Jianuo Huang 1,2* Yaojie Zhang 1,3* Qituan Zhang 4 Hao Lin 2 Hanlin Xu 5 Linfeng Zhang 1† 1 EPIC Lab, Shanghai Jiao Tong University 2 School of Software Engineering, HUST 3 UESTC 4 Fudan University 5 Huawei * Equal contribution. † Corresponding author |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 39749 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.29707