{"slug": "dominotree", "title": "DominoTree", "summary": "Researchers introduced DominoTree, a training-free best-first draft tree method for speculative decoding that uses Domino's conditional correction to achieve up to 6.6x speedup over autoregressive decoding on Qwen3-4B across eight benchmarks. The method constructs trees with a GPU-native CUDA-graph builder, outperforming existing drafters like DFlash and DDTree in throughput and acceptance length at various temperatures.", "body_md": "# Computer Science > Computation and Language\n\n[Submitted on 9 Jul 2026]\n\n# Title:DominoTree: Conditional Tree-Structured Drafting with Domino for Speculative Decoding\n\n[View PDF](/pdf/2607.08642)\n\n[HTML (experimental)](https://arxiv.org/html/2607.08642v1)\n\nAbstract:Speculative decoding accelerates LLM inference by drafting several tokens and verifying them in parallel. Block-diffusion drafters such as DFlash produce\n\na draft block in one pass but model only per-position marginals; best-first tree methods such as DDTree expand candidate trees from those marginals. The\n\nreleased Domino drafter adds a GRU-based causal correction that makes each draft token's distribution path-dependent, a structure DDTree's factorized\n\nformulation cannot represent. We introduce DominoTree, a training-free best-first draft tree scored by Domino's conditional, non-factorized correction\n\nalong each root-to-node path, made practical by restricting the per-node correction to a candidate top-M. On Qwen3-4B across eight benchmarks, DominoTree\n\nreaches up to 6.6x speedup over autoregressive decoding and the highest mean accept length of any evaluated method, up to 10.7 tokens per round, at every\n\ntemperature we test. DominoTree constructs its tree with a GPU-native, CUDA-graph builder that is bit-identical to a reference Python implementation, so\n\nacceptance is unchanged, while keeping per-round tree construction cheap. With this builder as default, DominoTree wins throughput over the released\n\nDomino decoder at every temperature, 9-10% overall on Qwen3-4B and up to +22% on Alpaca, and over DDTree/CaDDTree at every temperature we test. On Qwen3-\n\n8B, DominoTree keeps the highest accepted length at every temperature and adds a decisive throughput win at T=0, +24% over DDTree; at higher temperature\n\nthat edge over DDTree/CaDDTree narrows to a tie and a small loss, while its Overall aggregate wins over DFlash and Domino persist.\n\n### References & Citations\n\nLoading...\n\n# Bibliographic and Citation Tools\n\nBibliographic Explorer\n\n*(*[What is the Explorer?](https://info.arxiv.org/labs/showcase.html#arxiv-bibliographic-explorer))\nConnected Papers\n\n*(*[What is Connected Papers?](https://www.connectedpapers.com/about))\nLitmaps\n\n*(*[What is Litmaps?](https://www.litmaps.co/))\nscite Smart Citations\n\n*(*[What are Smart Citations?](https://www.scite.ai/))# Code, Data and Media Associated with this Article\n\nalphaXiv\n\n*(*[What is alphaXiv?](https://alphaxiv.org/))\nCatalyzeX Code Finder for Papers\n\n*(*[What is CatalyzeX?](https://www.catalyzex.com))\nDagsHub\n\n*(*[What is DagsHub?](https://dagshub.com/))\nGotit.pub\n\n*(*[What is GotitPub?](http://gotit.pub/faq))\nHugging Face\n\n*(*[What is Huggingface?](https://huggingface.co/huggingface))\nScienceCast\n\n*(*[What is ScienceCast?](https://sciencecast.org/welcome))# Demos\n\n# Recommenders and Search Tools\n\nInfluence Flower\n\n*(*[What are Influence Flowers?](https://influencemap.cmlab.dev/))\nCORE Recommender\n\n*(*[What is CORE?](https://core.ac.uk/services/recommender))# arXivLabs: experimental projects with community collaborators\n\narXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.\n\nBoth individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.\n\nHave an idea for a project that will add value for arXiv's community? [ Learn more about arXivLabs](https://info.arxiv.org/labs/index.html).", "url": "https://wpnews.pro/news/dominotree", "canonical_source": "https://arxiv.org/abs/2607.08642", "published_at": "2026-07-10 13:24:05+00:00", "updated_at": "2026-07-10 13:35:11.859030+00:00", "lang": "en", "topics": ["large-language-models", "machine-learning", "artificial-intelligence"], "entities": ["DominoTree", "DDTree", "DFlash", "Domino", "Qwen3-4B", "Qwen3-8B", "Alpaca", "CUDA"], "alternates": {"html": "https://wpnews.pro/news/dominotree", "markdown": "https://wpnews.pro/news/dominotree.md", "text": "https://wpnews.pro/news/dominotree.txt", "jsonld": "https://wpnews.pro/news/dominotree.jsonld"}}