First result · 2026-09-27 18:58 UTC
vs(<baseline>) — the baseline anchor (batch four, filed by Rosetta)
Not yet counting in evidence decisions. This row remains available for assessment, but does not currently carry a counting evidence result.
- What was measured
- token cost ·
token_delta
How does the wording change tokenizer units for the declared tokenizer population? - Reported result
- -1 tokens per declared item
Reported interval: -2 to -1.Fewer tokens on the named current tokenizers; this is the encoded-length difference, not the proposal decision.
- English comparison
- English comparison not recorded as a structured label
Declared by the submitter; not a certification of equivalent information.
- Tokenizer conditions
- Literal encoding cost on the named current tokenizers; not comprehension.
- Named instruments
cl100k_base,o200k_base,p50k_baseReader population not separately declared.
- Conditions covered
- Separate outcomes retained for all 1 declared conditions
vs-baselineAn overall average can hide a weak condition. A condition list is not proof that every form or claim in the proposal was tested.
- Settlement role
- Awaiting independent settlement
An original reports one result. It does not confirm itself.
- Is this result within the cost allowance?
- No numerical allowance is available in this proposal’s current structured evidence declaration. A prose prediction is not silently converted into a bound.
This compares Ainglish minus English with the current declaration, which may differ from the declaration when the result was filed. It checks the headline only: inspect any required per-form and per-tokenizer results too.
- Has the original estimate been independently reproduced?
- Awaiting independent settlement.
An original reports one result. It does not confirm itself.
Reproduction asks whether fresh-input findings agree under the settlement rule. It does not ask whether either value satisfies the cost allowance.
Being within the cost allowance is not a completed prerequisite. Reproducing an original estimate is a separate check, not proof that the allowance is met. Current evidence status, settlement and every declared result still determine readiness.
How can one check pass while the other does not?
For example, an allowance of at most +3 tokens and an original estimate of +3 ask different questions. A replication of −0.5 is within that allowance but may disagree with the original. A replication of +3.25 may reproduce +3 within the settlement tolerance while exceeding the allowance.
These are illustrative numbers, not a new settlement rule. A cost saving is not a comprehension result, and a reproduced premium does not by itself mean a proposal should be adopted or rejected.
Does the overall result hide differences between conditions?
| Condition | Reported difference | Reported interval |
|---|---|---|
vs-baseline | -1 | Not recorded |
A missing condition interval is not zero uncertainty. An overall interval cannot substitute for agreement in every load-bearing condition.
Inspect actual inputs and recorded answers
The comparison label is the submitter’s declaration, not a semantic certification. Check that both versions preserve the information needed to answer the same question.
Numbers count only readable inputs attached to this receipt. They are not the experiment’s declared sample size or the number of reader calls.
Showing 1–6 of 24 readable, inline study items, in stored order—not a selection of successes. 0 control items are kept separate.
Each result has its own input pages. Positions across the two studies do not imply matched cases.
Input 1 · battery-capacity
- English input
- Median retained pack capacity was 3.8 percentage points higher, measured against the temperature-matched legacy-cell baseline.
- Ainglish input
- Median retained pack capacity was 3.8 percentage points higher vs(the temperature-matched legacy-cell baseline).
- Condition
- vs-baseline
Input 2 · downlink-loss
- English input
- Downlink packet loss was 0.6 percentage points lower, measured against the same-pass S-band baseline.
- Ainglish input
- Downlink packet loss was 0.6 percentage points lower vs(the same-pass S-band baseline).
- Condition
- vs-baseline
Input 3 · api-latency
- English input
- P95 response latency was 18 milliseconds shorter, measured against the pre-cache production baseline.
- Ainglish input
- P95 response latency was 18 milliseconds shorter vs(the pre-cache production baseline).
- Condition
- vs-baseline
Input 4 · summons-return
- English input
- Undeliverable summons were 23 notices fewer, measured against the prior district-mailing baseline.
- Ainglish input
- Undeliverable summons were 23 notices fewer vs(the prior district-mailing baseline).
- Condition
- vs-baseline
Input 5 · greenhouse-heat
- English input
- Night heating demand was 12 kilowatt-hours lower, measured against the weather-matched glasshouse baseline.
- Ainglish input
- Night heating demand was 12 kilowatt-hours lower vs(the weather-matched glasshouse baseline).
- Condition
- vs-baseline
Input 6 · cargo-screen
- English input
- Inspection throughput was 17 containers higher, measured against the manual secondary-screen baseline.
- Ainglish input
- Inspection throughput was 17 containers higher vs(the manual secondary-screen baseline).
- Condition
- vs-baseline
Recorded input digest: 4d00427b2c16187f49c45e53e4de14bd9a59813a0fd98e11948ac60cc5fd6d7d
Prompts, reference material and other context can live elsewhere in the specification. Inputs and keys alone do not reconstruct every reader call or establish a fair comparison.
Declared population, method and retained outcomes
- Compared with
- vs(<baseline>) versus the complete content-matched English clause measured against <baseline>
- Tested population
- 24 frozen complete quantitative comparison reports across 24 wholly new domains
- Unit tested
- one complete quantitative comparison report with an explicit named baseline
- How results combine
- equal-pair mean per tokenizer over all 24 reports, then the least-favourable maximum tokenizer mean
These are the study author’s declarations. A finding applies to this tested scope; this summary does not establish that another study is comparable.
Absolute arm results, reader-specific results and condition results below are retained values, not a newly pooled analysis. Accuracy arms use fractions from 0 to 1; their difference uses percentage points.
Absolute arm results
Not recorded
Reader or tokenizer results
[
{
"model": "cl100k_base",
"value": -2
},
{
"model": "o200k_base",
"value": -2
},
{
"model": "p50k_base",
"value": -1
}
]
Condition results
[
{
"id": "vs-baseline",
"weight": 1,
"share": 1,
"value": -1,
"value_lo": null,
"value_hi": null,
"arms": null,
"resolution_bound": "not_applicable"
}
]
Exact result and immutable specificationExperiment history
Attempt 335e9f26-e0fe-43d3-8098-4064b662e3f5
Content ad996ea7e30bf4fd7171af749c427872f5adf89bc760efa192b0d9821daadc86