跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2611.21285 · v1

The Price of Not Planning: Parallel Sampling Depth Peaks at the Reconstruction Threshold

首次提交 2026-11-16 23:54
版本更新 2026-11-16 23:54
雷达观测 2026-11-17 16:10
arXiv ↗PDF ↗

摘要

Masked diffusion models sample the tokens of one round independently. Planned decoders first generate a coarse plan, then fill in the tokens in parallel. We ask when a plan saves rounds, how many it saves, and how fine it must be. We study this on broadcast trees: the leaves are the tokens, the internal nodes are the plan, and the edge correlation θθ sets how strongly tokens depend on the plan. Errors are measured by the forward Kullback-Leibler divergence ε\varepsilon, and the lower bounds hold for every deterministic value-adaptive rule, also with learned conditionals. We show that a decoder without the plan needs of order min⁡{n,(n/ε)α}\min\{n,(n/\varepsilon)^α\} rounds below the reconstruction threshold, where α=log⁡b/(4log⁡(1/θ))α=\log b/(4\log(1/θ)) rises to 1/21/2 at the threshold, and that the exponent stays below 1/21/2 above it (binary channel: 0.02≤θ≤0.980.02\leθ\le0.98). With the plan, log⁡bn+1\log_b n+1 rounds suffice. In the binary model below the threshold, a plan with units of mm tokens needs of order log⁡b(n/m)+min⁡{m,(n/ε)α}\log_b(n/m)+\min\{m,(n/\varepsilon)^α\} rounds, so it gives logarithmic depth exactly when mm is logarithmic. A plan therefore helps most near the threshold, and only if its units are small. Exact divergence computations in the Gaussian model with up to 2202^{20} leaves confirm these exponents below the threshold.

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

目前只收录了一个版本。收录下一版后,可以对照标题、作者、分类和摘要。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v1 · 2026-11-16 23:54 · 最新观测版本

The Price of Not Planning: Parallel Sampling Depth Peaks at the Reconstruction Threshold

Masked diffusion models sample the tokens of one round independently. Planned decoders first generate a coarse plan, then fill in the tokens in parallel. We ask when a plan saves rounds, how many it saves, and how fine it must be. We study this on broadcast trees: the leaves are the tokens, the internal nodes are the plan, and the edge correlation θθ sets how strongly tokens depend on the plan. Errors are measured by the forward Kullback-Leibler divergence ε\varepsilon, and the lower bounds hold for every deterministic value-adaptive rule, also with learned conditionals. We show that a decoder without the plan needs of order min⁡{n,(n/ε)α}\min\{n,(n/\varepsilon)^α\} rounds below the reconstruction threshold, where α=log⁡b/(4log⁡(1/θ))α=\log b/(4\log(1/θ)) rises to 1/21/2 at the threshold, and that the exponent stays below 1/21/2 above it (binary channel: 0.02≤θ≤0.980.02\leθ\le0.98). With the plan, log⁡bn+1\log_b n+1 rounds suffice. In the binary model below the threshold, a plan with units of mm tokens needs of order log⁡b(n/m)+min⁡{m,(n/ε)α}\log_b(n/m)+\min\{m,(n/\varepsilon)^α\} rounds, so it gives logarithmic depth exactly when mm is logarithmic. A plan therefore helps most near the threshold, and only if its units are small. Exact divergence computations in the Gaussian model with up to 2202^{20} leaves confirm these exponents below the threshold.

查看此版本解读 →查看此版本原文 ↗