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

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.

Question	First result	Second result
Which proposal?	cause-question(<E>) / justification-question(<A>) — did ‘why?’ ask what produced it, or what made it warranted?2026-09-02 21:38 UTC	cause-question(<E>) / justification-question(<A>) — did ‘why?’ ask what produced it, or what made it warranted?2026-09-13 10:35 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?	-11.6075 percentage pointsReported interval: -17.3977 to -5.9387Not yet counting in evidence decisions	-25.0025 percentage pointsReported interval: -42.5 to -11.1111Counts in current evidence decisions
Compared with what?	Other declared comparison; inspect the specificationReader exposure not recorded as a structured label	Other declared comparison; inspect the specificationReader 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 result Inspect 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

[
    "committed-per-item-comparator-v1"
]

Second result

[
    "committed-per-item-comparator-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) · Same recorded value

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

[
    "mistral-small3.2-24b-opaque-choice-q4_k_m@q4_k_m",
    "gemma3-12b-opaque-choice-q4_k_m@q4_k_m"
]

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": "cause-question-careful",
        "weight": 1
    },
    {
        "id": "cause-question-bare",
        "weight": 1
    },
    {
        "id": "justification-question-careful",
        "weight": 1
    },
    {
        "id": "justification-question-bare",
        "weight": 1
    }
]

Second result

[
    {
        "id": "cause-question-careful",
        "weight": 1
    },
    {
        "id": "cause-question-bare",
        "weight": 1
    },
    {
        "id": "justification-question-careful",
        "weight": 1
    },
    {
        "id": "justification-question-bare",
        "weight": 1
    }
]

First result: full recordcomprehension accuracy · -11.6075 percentage points

First result · 2026-09-02 21:38 UTC

cause-question(<E>) / justification-question(<A>) — did ‘why?’ ask what produced it, or what made it warranted?

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

-11.6075 percentage points
Reported interval: -17.3977 to -5.9387.

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

English comparison

Other declared comparison; inspect the specification

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

Careful-English strata estimate non-inferiority at -5 percentage points; balanced bare-why strata estimate recovery of the otherwise hidden causal-versus-normative relation.

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 4 declared conditions

cause-question-careful, cause-question-bare, justification-question-careful, justification-question-bare

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: 95.83%; 95.83%
Ainglish version: 84.22%; 84.22%

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: -17.3977 to -5.9387 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: 160 · 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.
Condition	Reported difference	Reported interval	English accuracy	Ainglish accuracy
`cause-question-careful`	-24.39	Not recorded	100.00%	75.61%
`cause-question-bare`	-9.76	Not recorded	100.00%	90.24%
`justification-question-careful`	0	Not recorded	100.00%	100.00%
`justification-question-bare`	-12.28	Not recorded	83.33%	71.05%

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: dd0b2673bd8b4943158daf6496f71fab280f4e2eaaf9dee0cf44a228af465691. 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": 0.95830000000000004067857162226573564112186431884765625,
    "ainglish": 0.84219999999999994866328734133276157081127166748046875,
    "chance": 0.25
}

Reader or tokenizer results

[
    {
        "model": "mistral-small3.2-24b-opaque-choice-q4_k_m",
        "value": -7.67250000000000031974423109204508364200592041015625,
        "precision": "q4_k_m"
    },
    {
        "model": "gemma3-12b-opaque-choice-q4_k_m",
        "value": -12.5675000000000007815970093361102044582366943359375,
        "precision": "q4_k_m"
    }
]

Condition results

[
    {
        "id": "cause-question-careful",
        "weight": 1,
        "share": 0.25,
        "value": -24.3900000000000005684341886080801486968994140625,
        "value_lo": null,
        "value_hi": null,
        "arms": {
            "english": 1,
            "ainglish": 0.756099999999999994315658113919198513031005859375,
            "chance": 0.25
        },
        "resolution_bound": "resolvable"
    },
    {
        "id": "cause-question-bare",
        "weight": 1,
        "share": 0.25,
        "value": -9.7599999999999997868371792719699442386627197265625,
        "value_lo": null,
        "value_hi": null,
        "arms": {
            "english": 1,
            "ainglish": 0.90239999999999997992716771477716974914073944091796875,
            "chance": 0.25
        },
        "resolution_bound": "ceiling"
    },
    {
        "id": "justification-question-careful",
        "weight": 1,
        "share": 0.25,
        "value": 0,
        "value_lo": null,
        "value_hi": null,
        "arms": {
            "english": 1,
            "ainglish": 1,
            "chance": 0.25
        },
        "resolution_bound": "ceiling"
    },
    {
        "id": "justification-question-bare",
        "weight": 1,
        "share": 0.25,
        "value": -12.2799999999999993605115378159098327159881591796875,
        "value_lo": null,
        "value_hi": null,
        "arms": {
            "english": 0.83330000000000004067857162226573564112186431884765625,
            "ainglish": 0.71050000000000002042810365310288034379482269287109375,
            "chance": 0.25
        },
        "resolution_bound": "resolvable"
    }
]

Exact result and immutable specification Experiment history

Attempt 160a343e-753a-425c-8e5f-969f84b22c3a
Content 4c90793b0dac00fb8ac214057ade4e5f80552cf484dad1829ed239331e9b1586

Second result: full recordcomprehension accuracy · -25.0025 percentage points

Second result · 2026-09-13 10:35 UTC

cause-question(<E>) / justification-question(<A>) — did ‘why?’ ask what produced it, or what made it warranted?

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

-25.0025 percentage points
Reported interval: -42.5 to -11.1111.

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

English comparison

Other declared comparison; inspect the specification

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

Careful-English strata estimate non-inferiority at -5 percentage points; balanced bare-why strata estimate recovery of the otherwise hidden causal-versus-normative relation.

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 4 declared conditions

cause-question-careful, cause-question-bare, justification-question-careful, justification-question-bare

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.

How much input text was reused?

Complete-pair freshness is not available for this receipt.

Separate-arm overlap is unavailable or has not been computed. This does not mean zero reuse.

Exact text comparisons only; repeated occurrences count separately. Shared text can deserve scrutiny even when each complete pair is new. These arm counts are descriptive and do not change settlement eligibility.

Declared target content identity

4c90793b0dac00fb8ac214057ade4e5f80552cf484dad1829ed239331e9b1586

How often did each version lead to the right answer?

English comparison: 66.67%; 66.67%
Ainglish version: 41.67%; 41.67%

Reported item-bootstrap interval: -42.5 to -11.1111 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.

Item-selection sensitivity warning. At least one reported reduced-item check changed direction or fell outside the full-item interval. Keep this warning with the score: the headline interval alone does not resolve sensitivity to which cases were included.

Real cases: 24 · 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.
Condition	Reported difference	Reported interval	English accuracy	Ainglish accuracy
`cause-question-careful`	-100	Not recorded	100.00%	0.00%
`cause-question-bare`	0	Not recorded	0.00%	0.00%
`justification-question-careful`	-16.67	Not recorded	100.00%	83.33%
`justification-question-bare`	16.66	Not recorded	66.67%	83.33%

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. 4 control items are kept separate.

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