Forcing the RL-trained curator to receive no retrieved trajectories degrades it to or below untrained JITMEM-base across all executors, with maximum reported success-rate drops of 14.8 points on ALFWorld and 15.2 on WebShop; the paper takes this as evidence that RL learns to distill retrieved experience rather than only emit hints from parametric knowledge.

Forcing the RL-trained curator to receive no retrieved trajectories degrades it to or below untrained JITMEM-base across all executors, with maximum reported success-rate drops of 14.8 points on ALFWorld and 15.2 on WebShop; the paper takes this as evidence that RL learns to distill retrieved experience rather than only emit hints from parametric knowledge. · CiteArk

Measurement results

0/9 assessed

0 supported0 disputed or conflicting9 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
62.6 percentage points	—	Not assessed
47.9 points	—	Not assessed
17.6 percentage points	—	Not assessed
79 percentage points	—	Not assessed
49.6 points	—	Not assessed
40.7 percentage points	—	Not assessed
72.9 percentage points	—	Not assessed
50 points	—	Not assessed
40.9 percentage points	—	Not assessed

Reported and observed measurements

Assessments (0)

Runs (0)

Measurement results