Against a two-pass hindsight scheduling oracle on the same mixed-forward backend, ASPIRE was 4.7–20.2% below oracle throughput, with 10.4% mean gap; the three long-context cells averaged 8.5%. The widest gap was Qwen3-1.7B on AIME25, and all other configurations were within 11.9%.

Against a two-pass hindsight scheduling oracle on the same mixed-forward backend, ASPIRE was 4.7–20.2% below oracle throughput, with 10.4% mean gap; the three long-context cells averaged 8.5%. The widest gap was Qwen3-1.7B on AIME25, and all other configurations were within 11.9%. · CiteArk

测量结果

已评估 0/26 项

获得支持 0有争议或冲突 0尚未评估 26

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
model: Qwen3-1.7B · dataset: AIME25	42 requests	—	尚未评估
model: Qwen3-1.7B · dataset: AIME25	2.59 × AutoRegressive	—	尚未评估
model: Qwen3-1.7B · dataset: AIME25	2.06 × AutoRegressive	—	尚未评估
model: Qwen3-1.7B · dataset: AIME25	20.2%	—	尚未评估
model: Qwen3-1.7B · dataset: LB-v2[30k-40k]	18 requests	—	尚未评估
model: Qwen3-1.7B · dataset: LB-v2[30k-40k]	4.46 × AutoRegressive	—	尚未评估
model: Qwen3-1.7B · dataset: LB-v2[30k-40k]	4.17 × AutoRegressive	—	尚未评估
model: Qwen3-1.7B · dataset: LB-v2[30k-40k]	6.5%	—	尚未评估
model: Qwen3-8B · dataset: AIME25	30 requests	—	尚未评估
model: Qwen3-8B · dataset: AIME25	1.95 × AutoRegressive	—	尚未评估
model: Qwen3-8B · dataset: AIME25	1.86 × AutoRegressive	—	尚未评估
model: Qwen3-8B · dataset: AIME25	4.7%	—	尚未评估

Against a two-pass hindsight scheduling oracle on the same mixed-forward backend, ASPIRE was 4.7–20.2% below oracle throughput, with 10.4% mean gap; the three long-context cells averaged 8.5%. The widest gap was Qwen3-1.7B on AIME25, and all other configurations were within 11.9%.

报告指标与观测值

Assessment（0）

关联运行（0）

测量结果