On ALFWorld with GPT-5.4 executor, JITMEM uses 9.8K input tokens, 0.87K output tokens, and 11.6 steps per task; relative to JITMEM-base, the paper reports reductions of 10.1%, 13.0%, and 12.1%, while all memory methods add input context but generally reduce output tokens and steps versus no memory.

On ALFWorld with GPT-5.4 executor, JITMEM uses 9.8K input tokens, 0.87K output tokens, and 11.6 steps per task; relative to JITMEM-base, the paper reports reductions of 10.1%, 13.0%, and 12.1%, while all memory methods add input context but generally reduce output tokens and steps versus no memory. · CiteArk

测量结果

已评估 0/24 项

获得支持 0有争议或冲突 0尚未评估 24

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。
论文报告	最近可用实测	证据判断
9 thousand tokens per task	—	尚未评估
1.4 thousand tokens per task	—	尚未评估
17.8 steps per task	—	尚未评估
19.7 thousand tokens per task	—	尚未评估
1.26 thousand tokens per task	—	尚未评估
16.2 steps per task	—	尚未评估
22.4 thousand tokens per task	—	尚未评估
1.36 thousand tokens per task	—	尚未评估
16.9 steps per task	—	尚未评估
10.9 thousand tokens per task	—	尚未评估
1 thousand tokens per task	—	尚未评估
13.2 steps per task	—	尚未评估

On ALFWorld with GPT-5.4 executor, JITMEM uses 9.8K input tokens, 0.87K output tokens, and 11.6 steps per task; relative to JITMEM-base, the paper reports reductions of 10.1%, 13.0%, and 12.1%, while all memory methods add input context but generally reduce output tokens and steps versus no memory.

报告指标与观测值

Assessment（0）

关联运行（0）

测量结果