On Qwen3-8B/LB[16k-18k], neither the smoothing-weight nor acceptance-prior sweep was monotone and no tested setting collapsed throughput. The paper says spread within each sweep was comparable to repeated-run variation and retained ω=0.8 and α₀=0.9 for reported results.

On Qwen3-8B/LB[16k-18k], neither the smoothing-weight nor acceptance-prior sweep was monotone and no tested setting collapsed throughput. The paper says spread within each sweep was comparable to repeated-run variation and retained ω=0.8 and α₀=0.9 for reported results. · CiteArk

测量结果

已评估 0/10 项

获得支持 0有争议或冲突 0尚未评估 10

保留同一结论下的完整结果组。已有评估排在前面；单项判断不替代整组结论。展开行查看条件、来源与历次证据。

每行是一项测量，分别显示论文报告、最近可用实测和累计证据状态。
指标与条件	论文报告	最近可用实测	证据判断
parameterValue: 0.4 · sweptParameter: smoothing weight ω	1,650 output token/s	—	尚未评估
parameterValue: 0.4 · sweptParameter: acceptance prior α₀	1,429 output token/s	—	尚未评估
parameterValue: 0.6 · sweptParameter: smoothing weight ω	1,599 output token/s	—	尚未评估
parameterValue: 0.6 · sweptParameter: acceptance prior α₀	1,419 output token/s	—	尚未评估
parameterValue: 0.8 · sweptParameter: smoothing weight ω	1,501 output token/s	—	尚未评估
parameterValue: 0.8 · sweptParameter: acceptance prior α₀	1,582 output token/s	—	尚未评估
parameterValue: 0.9 · sweptParameter: smoothing weight ω	1,731 output token/s	—	尚未评估
parameterValue: 0.9 · sweptParameter: acceptance prior α₀	1,501 output token/s	—	尚未评估
parameterValue: 0.95 · sweptParameter: smoothing weight ω	1,628 output token/s	—	尚未评估
parameterValue: 0.95 · sweptParameter: acceptance prior α₀	1,598 output token/s	—	尚未评估

On Qwen3-8B/LB[16k-18k], neither the smoothing-weight nor acceptance-prior sweep was monotone and no tested setting collapsed throughput. The paper says spread within each sweep was comparable to repeated-run variation and retained ω=0.8 and α₀=0.9 for reported results.

报告指标与观测值

Assessment（0）

关联运行（0）

测量结果