Ainglish An English dialect for AI agents
Menu

Evidence

Compare two experiments

Check what changed in the question, wording or readers before comparing the numbers.

This view keeps both results separate. It does not calculate a combined score or decide whether they reproduce each other.

Change the selected experimentsSearch by proposal, select a result or enter an exact identifier

Find experiments by proposal

Search for ordinary words from a proposal, then choose a match. Searching alone does not change the results below.

Choose two experiments by title, measurement and date. The choices include up to 50 newest public completed results, plus your current selections. Historical results stay labelled. For older records, use the evidence explorer or exact entry below.

Use an exact experiment identifier instead

A full attempt UUID entered here replaces the corresponding choice above. Content hashes are not result identifiers.

Clear comparison

Compare the essentials

Read each question across both results. A matching description is not proof that the studies are scientifically comparable.

QuestionFirst resultSecond result
Which proposal?set-to / adjust-by — is the number the new value, or the size of the change?2026-09-05 22:06 UTCset-to / adjust-by — is the number the new value, or the size of the change?2026-09-11 08:50 UTC
What was measured?comprehension accuracyHow does the wording change correct answers from the declared reader panel?comprehension accuracyHow does the wording change correct answers from the declared reader panel?
What was the result?-33.4683 percentage pointsReported interval: -43.056 to -23.3783Not yet counting in evidence decisions0 percentage pointsReported interval: 0 to 0Counts in current evidence decisions
Compared with what?Complete, careful EnglishReader exposure not recorded as a structured labelComplete, careful EnglishReader exposure not recorded as a structured label
Independently settled?DisputedEligible replications disagree and this original does not hold a settlement majority.Disagrees with the named originalThis eligible row adds one disagreement. An adverse or null direction is a valid result and remains visible.

Inspect first resultInspect second result

What differs in the declared methods?

These are literal recorded values, not a compatibility test. Matching declarations do not prove equivalent inputs or fair scoring. A missing structured field may be described elsewhere in the immutable specification.

What was measured · Same recorded value
First result
comprehension_accuracy_delta
Second result
comprehension_accuracy_delta
English comparison declarations · Same recorded value
First result
[
    "complete-careful-english-v1"
]
Second result
[
    "complete-careful-english-v1"
]
Tested population · Not recorded on either side
First result
Not recorded in this structured field
Second result
Not recorded in this structured field
Unit tested · Not recorded on either side
First result
Not recorded in this structured field
Second result
Not recorded in this structured field
Named instruments (including order) · Recorded values differ
First result
[
    "mistral-small3.2-24b-opaque-choice-q4_k_m@q4_k_m",
    "gemma3-12b-opaque-choice-q4_k_m@q4_k_m"
]
Second result
[
    "spark-cli-13"
]
Reader exposure · Not recorded on either side
First result
Not recorded in this structured field
Second result
Not recorded in this structured field
Exposure window · Not recorded on either side
First result
Not recorded in this structured field
Second result
Not recorded in this structured field
How results combine · Not recorded on either side
First result
Not recorded in this structured field
Second result
Not recorded in this structured field
Scoring declaration · Not recorded on either side
First result
Not recorded in this structured field
Second result
Not recorded in this structured field
Conditions (including order) · Same recorded value
First result
[
    {
        "id": "set-to:known",
        "weight": 1
    },
    {
        "id": "adjust-by:known",
        "weight": 1
    },
    {
        "id": "set-to:unknown",
        "weight": 1
    },
    {
        "id": "adjust-by:unknown",
        "weight": 1
    },
    {
        "id": "set-to:ordered",
        "weight": 1
    },
    {
        "id": "adjust-by:ordered",
        "weight": 1
    }
]
Second result
[
    {
        "id": "set-to:known",
        "weight": 1
    },
    {
        "id": "adjust-by:known",
        "weight": 1
    },
    {
        "id": "set-to:unknown",
        "weight": 1
    },
    {
        "id": "adjust-by:unknown",
        "weight": 1
    },
    {
        "id": "set-to:ordered",
        "weight": 1
    },
    {
        "id": "adjust-by:ordered",
        "weight": 1
    }
]
First result: full recordcomprehension accuracy · -33.4683 percentage points

First result · 2026-09-05 22:06 UTC

set-to / adjust-by — is the number the new value, or the size of the change?

Not yet counting in evidence decisions. This row remains available for assessment, but does not currently carry a counting evidence result.

What was measured
comprehension accuracy · comprehension_accuracy_delta
How does the wording change correct answers from the declared reader panel?
Reported result
-33.4683 percentage points
Reported interval: -43.056 to -23.3783.

The value falls on the registered harmful side of this metric’s neutral point.

English comparison
Complete, careful English

Declared by the submitter; not a certification of equivalent information.

Identical contextual facts in both arms; direct complete English for the question asked. Scope and omitted dimensions are explicit in DESIGN.md.

Reader exposure
Reader exposure not recorded as a structured label
Named instruments
mistral-small3.2-24b-opaque-choice-q4_k_m@q4_k_m, gemma3-12b-opaque-choice-q4_k_m@q4_k_m

Reader population not separately declared.

Conditions covered
Separate outcomes retained for all 6 declared conditions

set-to:known, adjust-by:known, set-to:unknown, adjust-by:unknown, set-to:ordered, adjust-by:ordered

An 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
Disputed

Eligible replications disagree and this original does not hold a settlement majority.

How often did each version lead to the right answer?

English comparison
91.95%
91.95%
Ainglish version
58.48%
58.48%

Reported real-item accuracy, not the separate calibration score. Both bars use the same 0–100% scale. The difference is measured in percentage points, not percent improvement. Any declared stratum weights are already applied.

Reported item-bootstrap interval: -43.056 to -23.3783 percentage points.

This interval concerns the difference, not separate uncertainty bounds for either accuracy. It does not measure uncertainty across humans or future models.

At least one declared condition is resolution-limited. The overall interval does not settle every condition.

Real cases: 96 · Named readers: 2. These are different units; multiple answers to one case are not new cases.

Does the overall result hide differences between conditions?

Every stored condition, without new pooling. Differences and intervals use percentage points. Condition names come from the frozen experiment.
ConditionReported differenceReported intervalEnglish accuracyAinglish accuracy
set-to:known-75 Not recorded 100.00%25.00%
adjust-by:known-7.14 Not recorded 100.00%92.86%
set-to:unknown-61.11 Not recorded 100.00%38.89%
adjust-by:unknown8.33 Not recorded 91.67%100.00%
set-to:ordered-64.71 Not recorded 100.00%35.29%
adjust-by:ordered-1.18 Not recorded 60.00%58.82%

A missing condition interval is not zero uncertainty. An overall interval cannot substitute for agreement in every load-bearing condition.

Inspect externally stored 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.

The input material is linked externally. The number of study items and controls in that file has not been checked by this website. “External file” does not mean zero inputs.

Open the declared external input artifact. This is an unverified external link, not a hosted or inspected copy.

Declared input digest: 8f29af0b04a02f2502dc9483bb3e929687429c6e8eaf59b0b7403cd644fbdeb7. A recorded digest alone does not establish that the linked file matches it.

The website does not fetch the file. Verify the declared digest recipe before relying on it: SDK item digests use canonical JSON of the item array, not the raw pretty-printed file bytes.

No readable study input pairs are stored inline in this receipt. This does not mean the experiment used none.

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

No structured study scope is declared here. Inspect the immutable manifest; do not infer a comparator or population from the headline.

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

{
    "english": 0.9194999999999999840127884453977458178997039794921875,
    "ainglish": 0.584799999999999986499688020558096468448638916015625,
    "chance": 0.25
}

Reader or tokenizer results

[
    {
        "model": "mistral-small3.2-24b-opaque-choice-q4_k_m",
        "value": -40.873300000000000409272615797817707061767578125,
        "precision": "q4_k_m"
    },
    {
        "model": "gemma3-12b-opaque-choice-q4_k_m",
        "value": -28.121700000000000585487214266322553157806396484375,
        "precision": "q4_k_m"
    }
]

Condition results

[
    {
        "id": "set-to:known",
        "weight": 1,
        "share": 0.1666666666666666574148081281236954964697360992431640625,
        "value": -75,
        "value_lo": null,
        "value_hi": null,
        "arms": {
            "english": 1,
            "ainglish": 0.25,
            "chance": 0.25
        },
        "resolution_bound": "resolvable"
    },
    {
        "id": "adjust-by:known",
        "weight": 1,
        "share": 0.1666666666666666574148081281236954964697360992431640625,
        "value": -7.13999999999999968025576890795491635799407958984375,
        "value_lo": null,
        "value_hi": null,
        "arms": {
            "english": 1,
            "ainglish": 0.92859999999999998099298181841732002794742584228515625,
            "chance": 0.25
        },
        "resolution_bound": "ceiling"
    },
    {
        "id": "set-to:unknown",
        "weight": 1,
        "share": 0.1666666666666666574148081281236954964697360992431640625,
        "value": -61.1099999999999994315658113919198513031005859375,
        "value_lo": null,
        "value_hi": null,
        "arms": {
            "english": 1,
            "ainglish": 0.388900000000000023447910280083306133747100830078125,
            "chance": 0.25
        },
        "resolution_bound": "resolvable"
    },
    {
        "id": "adjust-by:unknown",
        "weight": 1,
        "share": 0.1666666666666666574148081281236954964697360992431640625,
        "value": 8.3300000000000000710542735760100185871124267578125,
        "value_lo": null,
        "value_hi": null,
        "arms": {
            "english": 0.91669999999999995932142837773426435887813568115234375,
            "ainglish": 1,
            "chance": 0.25
        },
        "resolution_bound": "ceiling"
    },
    {
        "id": "set-to:ordered",
        "weight": 1,
        "share": 0.1666666666666666574148081281236954964697360992431640625,
        "value": -64.7099999999999937472239253111183643341064453125,
        "value_lo": null,
        "value_hi": null,
        "arms": {
            "english": 1,
            "ainglish": 0.3528999999999999914734871708787977695465087890625,
            "chance": 0.25
        },
        "resolution_bound": "resolvable"
    },
    {
        "id": "adjust-by:ordered",
        "weight": 1,
        "share": 0.1666666666666666574148081281236954964697360992431640625,
        "value": -1.1799999999999999378275106209912337362766265869140625,
        "value_lo": null,
        "value_hi": null,
        "arms": {
            "english": 0.59999999999999997779553950749686919152736663818359375,
            "ainglish": 0.58819999999999994511057366253226064145565032958984375,
            "chance": 0.25
        },
        "resolution_bound": "resolvable"
    }
]

Exact result and immutable specificationExperiment history

Attempt 00910c7b-a19f-4533-8c94-688cdc326a2c
Content e7b399a86856b1e31f5c9afdb92fea761a150698c8acab24c24c224d6a8d1b44

Second result: full recordcomprehension accuracy · 0 percentage points

Second result · 2026-09-11 08:50 UTC

set-to / adjust-by — is the number the new value, or the size of the change?

Counts in current evidence decisions. This row currently contributes to evidence decisions. Its direction is separate from whether the proposal is ready for adoption.

What was measured
comprehension accuracy · comprehension_accuracy_delta
How does the wording change correct answers from the declared reader panel?
Reported result
0 percentage points
Reported interval: 0 to 0.

The value is neutral or does not resolve the registered direction.

English comparison
Complete, careful English

Declared by the submitter; not a certification of equivalent information.

Identical contextual facts in both arms; direct complete English for the question asked. Scope explicit per item in the frozen set.

Reader exposure
Reader exposure not recorded as a structured label
Named instruments
spark-cli-13

Reader population not separately declared.

Conditions covered
Separate outcomes retained for all 6 declared conditions

set-to:known, adjust-by:known, set-to:unknown, adjust-by:unknown, set-to:ordered, adjust-by:ordered

An 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
Disagrees with the named original

This eligible row adds one disagreement. An adverse or null direction is a valid result and remains visible.

This result checks a named original, not every experiment on the proposal. Read its target original

Compare with the exact target attempt

Every declared condition must agree. Overlapping overall intervals alone do not confirm this original.

Declared target content identitye7b399a86856b1e31f5c9afdb92fea761a150698c8acab24c24c224d6a8d1b44

How often did each version lead to the right answer?

English comparison
100.00%
100.00%
Ainglish version
100.00%
100.00%

Reported real-item accuracy, not the separate calibration score. Both bars use the same 0–100% scale. The difference is measured in percentage points, not percent improvement. Any declared stratum weights are already applied.

Reported item-bootstrap interval: 0 to 0 percentage points.

This interval concerns the difference, not separate uncertainty bounds for either accuracy. It does not measure uncertainty across humans or future models.

At least one declared condition is resolution-limited. The overall interval does not settle every condition.

Real cases: 16 · Named readers: 1. These are different units; multiple answers to one case are not new cases.

Does the overall result hide differences between conditions?

Every stored condition, without new pooling. Differences and intervals use percentage points. Condition names come from the frozen experiment.
ConditionReported differenceReported intervalEnglish accuracyAinglish accuracy
set-to:known0 Not recorded 100.00%100.00%
adjust-by:known0 Not recorded 100.00%100.00%
set-to:unknown0 Not recorded 100.00%100.00%
adjust-by:unknown0 Not recorded 100.00%100.00%
set-to:ordered0 Not recorded 100.00%100.00%
adjust-by:ordered0 Not recorded 100.00%100.00%

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 16 readable, inline study items, in stored order—not a selection of successes. 3 control items are kept separate.

Each result has its own input pages. Positions across the two studies do not imply matched cases.

Input 4 · real-s1

English input
One scalar quantity is in scope: Greenhouse G-7 humidity setpoint, measured in pct. Its starting value is 45 pct. Set this quantity to 60 pct.
Ainglish input
One scalar quantity is in scope: Greenhouse G-7 humidity setpoint, measured in pct. Its starting value is 45 pct. Greenhouse G-7 humidity setpoint set-to(+60 pct).
Question
Assuming the stated updates complete exactly in their stated order, what is the final value? Answer with the option letter only. A = 45 pct. B = 60 pct. C = 105 pct. D = not determined.
Recorded answer options
A · B · C · D
Submitted answer key
B. This is the supplied key, not an independent validation of it.
Condition
set-to:known

Filed correctness against that key (up to 12 reader/arm cells). These flags are not the reader’s verbatim output.

  • spark-cli-13 · Ainglish: matched the submitted key.

Input 5 · real-s2

English input
One scalar quantity is in scope: Kiln K-3 firing temperature, measured in degC. Its starting value is 800 degC. Set this quantity to 950 degC.
Ainglish input
One scalar quantity is in scope: Kiln K-3 firing temperature, measured in degC. Its starting value is 800 degC. Kiln K-3 firing temperature set-to(+950 degC).
Question
Assuming the stated updates complete exactly in their stated order, what is the final value? Answer with the option letter only. A = not determined. B = 1750 degC. C = 950 degC. D = 800 degC.
Recorded answer options
A · B · C · D
Submitted answer key
C. This is the supplied key, not an independent validation of it.
Condition
set-to:known

Filed correctness against that key (up to 12 reader/arm cells). These flags are not the reader’s verbatim output.

  • spark-cli-13 · English: matched the submitted key.

Input 6 · real-s3

English input
One scalar quantity is in scope: Reservoir R-9 outflow gate, measured in cm. Its starting value is 30 cm. Set this quantity to 12 cm.
Ainglish input
One scalar quantity is in scope: Reservoir R-9 outflow gate, measured in cm. Its starting value is 30 cm. Reservoir R-9 outflow gate set-to(+12 cm).
Question
Assuming the stated updates complete exactly in their stated order, what is the final value? Answer with the option letter only. A = not determined. B = 12 cm. C = 42 cm. D = 30 cm.
Recorded answer options
A · B · C · D
Submitted answer key
B. This is the supplied key, not an independent validation of it.
Condition
set-to:known

Filed correctness against that key (up to 12 reader/arm cells). These flags are not the reader’s verbatim output.

  • spark-cli-13 · Ainglish: matched the submitted key.

Input 7 · real-s4

English input
One scalar quantity is in scope: Cleanroom C-2 pressure offset, measured in Pa. Its starting value is -5 Pa. Set this quantity to 8 Pa.
Ainglish input
One scalar quantity is in scope: Cleanroom C-2 pressure offset, measured in Pa. Its starting value is -5 Pa. Cleanroom C-2 pressure offset set-to(+8 Pa).
Question
Assuming the stated updates complete exactly in their stated order, what is the final value? Answer with the option letter only. A = 3 Pa. B = not determined. C = 8 Pa. D = -5 Pa.
Recorded answer options
A · B · C · D
Submitted answer key
C. This is the supplied key, not an independent validation of it.
Condition
set-to:known

Filed correctness against that key (up to 12 reader/arm cells). These flags are not the reader’s verbatim output.

  • spark-cli-13 · English: matched the submitted key.

Input 8 · real-a1

English input
One scalar quantity is in scope: Silo S-4 grain reserve, measured in t. Its starting value is 120 t. Adjust this quantity by +35 t.
Ainglish input
One scalar quantity is in scope: Silo S-4 grain reserve, measured in t. Its starting value is 120 t. Silo S-4 grain reserve adjust-by(+35 t).
Question
Assuming the stated updates complete exactly in their stated order, what is the final value? Answer with the option letter only. A = 155 t. B = 120 t. C = 85 t. D = not determined.
Recorded answer options
A · B · C · D
Submitted answer key
A. This is the supplied key, not an independent validation of it.
Condition
adjust-by:known

Filed correctness against that key (up to 12 reader/arm cells). These flags are not the reader’s verbatim output.

  • spark-cli-13 · English: matched the submitted key.

Input 9 · real-a2

English input
One scalar quantity is in scope: Battery B-6 backup charge, measured in pct. Its starting value is 70 pct. Adjust this quantity by -25 pct.
Ainglish input
One scalar quantity is in scope: Battery B-6 backup charge, measured in pct. Its starting value is 70 pct. Battery B-6 backup charge adjust-by(-25 pct).
Question
Assuming the stated updates complete exactly in their stated order, what is the final value? Answer with the option letter only. A = not determined. B = 70 pct. C = 45 pct. D = 95 pct.
Recorded answer options
A · B · C · D
Submitted answer key
C. This is the supplied key, not an independent validation of it.
Condition
adjust-by:known

Filed correctness against that key (up to 12 reader/arm cells). These flags are not the reader’s verbatim output.

  • spark-cli-13 · Ainglish: matched the submitted key.

Recorded input digest: d2e02af1395c3a81308d49151cc8dfa8fb42b6bdfe9589c85f6f36f5a93156fb

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

No structured study scope is declared here. Inspect the immutable manifest; do not infer a comparator or population from the headline.

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

{
    "english": 1,
    "ainglish": 1,
    "chance": 0.25
}

Reader or tokenizer results

[
    {
        "model": "spark-cli-13",
        "value": 0
    }
]

Condition results

[
    {
        "id": "set-to:known",
        "weight": 1,
        "share": 0.1666666666666666574148081281236954964697360992431640625,
        "value": 0,
        "value_lo": null,
        "value_hi": null,
        "arms": {
            "english": 1,
            "ainglish": 1,
            "chance": 0.25
        },
        "resolution_bound": "ceiling"
    },
    {
        "id": "adjust-by:known",
        "weight": 1,
        "share": 0.1666666666666666574148081281236954964697360992431640625,
        "value": 0,
        "value_lo": null,
        "value_hi": null,
        "arms": {
            "english": 1,
            "ainglish": 1,
            "chance": 0.25
        },
        "resolution_bound": "ceiling"
    },
    {
        "id": "set-to:unknown",
        "weight": 1,
        "share": 0.1666666666666666574148081281236954964697360992431640625,
        "value": 0,
        "value_lo": null,
        "value_hi": null,
        "arms": {
            "english": 1,
            "ainglish": 1,
            "chance": 0.25
        },
        "resolution_bound": "ceiling"
    },
    {
        "id": "adjust-by:unknown",
        "weight": 1,
        "share": 0.1666666666666666574148081281236954964697360992431640625,
        "value": 0,
        "value_lo": null,
        "value_hi": null,
        "arms": {
            "english": 1,
            "ainglish": 1,
            "chance": 0.25
        },
        "resolution_bound": "ceiling"
    },
    {
        "id": "set-to:ordered",
        "weight": 1,
        "share": 0.1666666666666666574148081281236954964697360992431640625,
        "value": 0,
        "value_lo": null,
        "value_hi": null,
        "arms": {
            "english": 1,
            "ainglish": 1,
            "chance": 0.25
        },
        "resolution_bound": "ceiling"
    },
    {
        "id": "adjust-by:ordered",
        "weight": 1,
        "share": 0.1666666666666666574148081281236954964697360992431640625,
        "value": 0,
        "value_lo": null,
        "value_hi": null,
        "arms": {
            "english": 1,
            "ainglish": 1,
            "chance": 0.25
        },
        "resolution_bound": "ceiling"
    }
]

Exact result and immutable specificationExperiment history

Attempt 96e9b6ce-9684-49f6-877f-4d1feaedcc60
Content d246c98217b24a20537af312a0bcac3afd67b547336d4851c0b5b390ca8be719

Different wording, readers, exposure or populations can legitimately produce different results. A visible reference is not training the model’s weights. Current models and tokenizers have learned English; future Ainglish-trained performance remains a research question.