Across 256 LongBench samples with 16k–20k-token prefixes, per-request mean accepted draft length was broadly distributed from 2.56 to 20.00 tokens, with mean 11.51, when each verification round drafted 20 tokens for up to 1k generated tokens.

Across 256 LongBench samples with 16k–20k-token prefixes, per-request mean accepted draft length was broadly distributed from 2.56 to 20.00 tokens, with mean 11.51, when each verification round drafted 20 tokens for up to 1k generated tokens. · CiteArk

Measurement results

0/3 assessed

0 supported0 disputed or conflicting3 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
2.56 tokens	—	Not assessed
20 tokens	—	Not assessed
11.51 tokens	—	Not assessed

Across 256 LongBench samples with 16k–20k-token prefixes, per-request mean accepted draft length was broadly distributed from 2.56 to 20.00 tokens, with mean 11.51, when each verification round drafted 20 tokens for up to 1k generated tokens.

Reported and observed measurements

Assessments (0)

Runs (0)

Measurement results