Cost-Aware Diffusion Draft Trees for Speculative Decoding

• URL: http://arxiv.org/abs/2606.01813v1
• Date: Mon, 01 Jun 2026 07:29:56 GMT
• Title: Cost-Aware Diffusion Draft Trees for Speculative Decoding
• Abstract summary: We introduce textbfCaDDTree (Cost-aware Diffusion Draft Tree), a method that directly optimize token throughput (expected tokens generated per unit time) by jointly selecting the tree structure and node budget.<n>Experiments on Qwen3-4B and Qwen3-8B across eight benchmarks spanning reasoning, coding, and instruction-following tasks show that caDDTree matches or surpasses DDTree with oracle budget selection on nearly all tasks.
• Score: 17.900184934329666
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Speculative decoding accelerates inference by having a lightweight drafter propose tokens verified in parallel by the target language model. Block diffusion drafters such as DFlash generate an entire draft block in one pass, yielding per-position marginals; DDTree uses these to build a candidate tree that maximizes expected acceptance length under a fixed node budget. We observe, however, that acceptance length is non-decreasing in budget: it always favors larger trees regardless of verification cost, offering no principled basis for budget selection. We introduce \textbf{CaDDTree} (Cost-aware Diffusion Draft Tree), a method that directly optimizes token throughput (expected tokens generated per unit time) by jointly selecting the tree structure and node budget. We model draft and verification latencies explicitly, show that the throughput objective decomposes into a per-round one-dimensional search over the budget, and prove that under a convex verification cost the throughput function is \emph{unimodal}, enabling an efficient greedy stopping rule. CaDDTree requires no offline budget search, adapting the budget each round from the current per-position distributions and verification cost. Experiments on Qwen3-4B and Qwen3-8B across eight benchmarks spanning reasoning, coding, and instruction-following tasks show that \caDDTree{} matches or surpasses DDTree with oracle budget selection on nearly all tasks.

Related papers

• ECHO: Early-layer Collaborative Hierarchical Orchestration with Bonus Logits in Speculative Decoding [60.34475274870224]
  ECHO is a hierarchical dual-loop framework that exploits the functional asymmetry between LLM layers.<n>Within the inner loop, early-layer bonus logits drive rapid, multi-step draft-tree exploration at a minimal cost.<n>ECHO significantly boosts mean accepted tokens and achieves a 2.4$times$ to 2.9$times$ speedup.
  arXiv  Detail & Related papers  (2026-09-15T14:21:33Z)
• DARTree: Speculative Diffusion Decoding with Autoregressive Draft Trees [40.61855455828842]
  Speculative decoding losslessly accelerates autoregressive language models by verifying multiple draft tokens in parallel.<n>Dartree is a training-free speculative decoding method that extends a pretrained AR correction head from chains to trees.
  arXiv  Detail & Related papers  (2026-08-13T17:43:44Z)
• PRESTO: Prefix-Aligned Tree Drafting for Diffusion Speculative Decoding [98.71600061506206]
  We propose PRESTO, a principled framework that extends tree-based drafting to diffusion drafters.<n> PRESTO achieves up to an average of $1.5times$ end-to-end throughput speedup on the state-of-the-art dedicated diffusion drafter SD.
  arXiv  Detail & Related papers  (2026-06-20T14:32:14Z)
• Accelerating Speculative Decoding with Block Diffusion Draft Trees [20.28933257827737]
  Speculative decoding accelerates autoregressive language models by using a lightweight drafter to propose multiple future tokens.<n>DFlash shows that a block diffusion drafter can generate an entire draft block in a single forward pass.<n> Vanilla DFlash verifies only a single drafted trajectory per round, potentially limiting its acceptance length.<n>We introduce DDTree, a method that constructs a draft tree directly from the per-position distributions of a block diffusion drafter.
  arXiv  Detail & Related papers  (2026-04-14T17:23:14Z)
• Aligning Tree-Search Policies with Fixed Token Budgets in Test-Time Scaling of LLMs [16.21023293204606]
  Existing tree-search policies are largely budget-agnostic, treating the budget as a termination condition, which can lead to late-stage over-branching or premature termination.<n>We propose Budget-Guided MCTS (BG-MCTS), a tree-search decoding algorithm that aligns its search policy with the remaining token budget.<n>BG-MCTS consistently outperforms budget-agnostic tree-search baselines across different budgets on MATH500 and AIME24/25 with open-weight LLMs.
  arXiv  Detail & Related papers  (2026-02-10T09:23:26Z)
• TALON: Confidence-Aware Speculative Decoding with Adaptive Token Trees [18.53532655905144]
  Speculative decoding (SD) has become a standard technique for accelerating LLM inference without sacrificing output quality.<n>We introduce TALON, a training-free, budget-driven adaptive tree expansion framework that can be plugged into existing tree-based methods.<n> TALON consistently outperforms state-of-the-art Eagle-3, achieving up to 5.16x end-to-end speedup over auto-regressive decoding.
  arXiv  Detail & Related papers  (2026-01-12T09:26:45Z)
• Fast Inference of Visual Autoregressive Model with Adjacency-Adaptive Dynamical Draft Trees [50.230925890958936]
  We propose an adjacency-adaptive dynamic draft tree that adjusts draft tree depth and width by leveraging adjacent token states and prior acceptance rates.<n>ADT-Tree achieves speedups of 3.13xand 3.05x, respectively, and integrates seamlessly with relaxed sampling methods such as LANTERN.
  arXiv  Detail & Related papers  (2025-12-26T04:45:49Z)
• A*-Thought: Efficient Reasoning via Bidirectional Compression for Low-Resource Settings [60.48717743667377]
  A*-Thought is an efficient tree search-based unified framework designed to identify and isolate the most essential thoughts.<n>It formulates the reasoning process of LRMs as a search tree, where each node represents a reasoning span in the giant reasoning space.<n>It can improve the performance of QwQ-32B by 2.39$times$ with low-budget and reduce the length of the output token by nearly 50% with high-budget.
  arXiv  Detail & Related papers  (2025-05-30T12:58:34Z)
• Don't Get Lost in the Trees: Streamlining LLM Reasoning by Overcoming Tree Search Exploration Pitfalls [83.89771461061903]
  Recent advancements in tree search algorithms guided by verifiers have significantly enhanced the reasoning capabilities of large language models (LLMs)<n>Recent advancements in tree search algorithms guided by verifiers have significantly enhanced the reasoning capabilities of large language models (LLMs)<n>We identify two key challenges contributing to this inefficiency: $textitover-exploration$ due to redundant states with semantically equivalent content, and $textitunder-exploration$ caused by high variance in verifier scoring.<n>We propose FETCH, a flexible, plug-and-play system compatible with various tree search algorithms.
  arXiv  Detail & Related papers  (2025-02-16T16:12:01Z)
• OPT-Tree: Speculative Decoding with Adaptive Draft Tree Structure [40.9990864658776]
  Speculative decoding employs a "draft and then verify" mechanism to allow multiple tokens to be generated in one step.<n>Existing methods mainly adopt fixed draft structures, which fail to adapt to different situations.<n>We propose OPT-Tree, an algorithm to construct adaptive and scalable draft trees.
  arXiv  Detail & Related papers  (2024-06-25T04:45:53Z)
• On $\ell_p$-norm Robustness of Ensemble Stumps and Trees [83.81523991945018]
  We develop an efficient programming based algorithm for sound verification of ensemble stumps. We demonstrate the first certified defense method for training ensemble stumps and trees with respect to $ell_p$ norm perturbations.
  arXiv  Detail & Related papers  (2020-08-20T03:42:40Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.