On Qwen3-8B/LB[16k-18k], refresh and no-refresh variants were both approximately 95% acceptance at fixed draft length 1. At draft length 10, no-refresh acceptance fell to 60.3%, while refresh retained 76.6%; the paper reports the gap as 16.4 percentage points and says it widened steadily with draft length.

On Qwen3-8B/LB[16k-18k], refresh and no-refresh variants were both approximately 95% acceptance at fixed draft length 1. At draft length 10, no-refresh acceptance fell to 60.3%, while refresh retained 76.6%; the paper reports the gap as 16.4 percentage points and says it widened steadily with draft length. · CiteArk

Measurement results

0/5 assessed

0 supported0 disputed or conflicting5 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.
Metric and conditions	Reported	Latest observation	Evidence judgment
refresh: enabled · qualifier: approximately · fixedDraftLength: 1	95%	—	Not assessed
refresh: disabled · qualifier: approximately · fixedDraftLength: 1	95%	—	Not assessed
refresh: disabled · fixedDraftLength: 10	60.3%	—	Not assessed
refresh: enabled · fixedDraftLength: 10	76.6%	—	Not assessed
comparison: refresh minus no refresh · fixedDraftLength: 10	16.4 percentage points	—	Not assessed

Reported and observed measurements

Assessments (0)

Runs (0)

Measurement results