For untrained JITMEM-base, storing all trajectories with correctness labels rather than filtering to successful trajectories lowers success by 1.5-2.9 points on ALFWorld and 2.3-3.4 on WebShop across executors; replacing raw traces with write-time ReasoningBank-style distillations lowers success by 1.7-2.9 on ALFWorld and 6.8-8.2 on WebShop.

For untrained JITMEM-base, storing all trajectories with correctness labels rather than filtering to successful trajectories lowers success by 1.5-2.9 points on ALFWorld and 2.3-3.4 on WebShop across executors; replacing raw traces with write-time ReasoningBank-style distillations lowers success by 1.7-2.9 on ALFWorld and 6.8-8.2 on WebShop. · CiteArk

Measurement results

0/18 assessed

0 supported0 disputed or conflicting18 not assessed

The complete result group stays with its claim. Assessed rows come first; their judgments do not replace the whole-claim conclusion. Expand a row for conditions, sources, and evidence history.

Each row is a measurement with its reported value, latest available observation, and cumulative evidence state.
Reported	Latest observation	Evidence judgment
59 percentage points	—	Not assessed
23.5 points	—	Not assessed
8.9 percentage points	—	Not assessed
77.1 percentage points	—	Not assessed
52 points	—	Not assessed
42.1 percentage points	—	Not assessed
77.6 percentage points	—	Not assessed
45.5 points	—	Not assessed
35.9 percentage points	—	Not assessed
58.8 percentage points	—	Not assessed
17.8 points	—	Not assessed
4.9 percentage points	—	Not assessed

Reported and observed measurements

Assessments (0)

Runs (0)

Measurement results