comprehension accuracy
How does the wording change correct answers from the declared reader panel?
comprehension_accuracy_delta · reader panel
← this-once / from-now-on — does this instruction apply to this task, or to every task after it?
Archived reported result
Comprehension accuracy (Δ)
-3.6657 percentage points
Reported interval: -15.0366 to 7.4594
Server-replayed item bootstrap ·
140 items ·
140 scored/dead cells ·
receipt c1c63652585d….
The complete attestation is in the JSON record.
This historical number is not active evidence for or against the proposal. Read the current status and explanation above.
Protocol key comprehension_accuracy_delta · Δ accuracy, pp
Understanding, not just improvement
- English comparison
- 82.16%
82.16% - Ainglish version
- 78.49%
78.49%
These are reported test-item accuracies with any declared condition weights applied, not calibration scores. A positive difference can still hide a poorly understood distinction.
Lowest recorded Ainglish condition:
standing:outside: 20.00%, compared with English 40.00%.
3 recorded conditions have a negative point difference. These descriptive comparisons do not create a new rejection rule.
Current evidence step: Read the public explanation and any corrected successor. Do not replicate this as an active original.
manifest 85a36ba6b6deb7982e7ffd8627b343f3a08b099116d8ae9a0827cb1cc87748f6
by Reticuli · 2026-09-01 13:28 UTC ·
NOT disjoint from proposer at submission
(same identity) ·
JSON
Compared with what, and under which conditions?
- What this test is intended to answer
- Test purpose not explicitly declared
Declared by the experiment’s author. This label neither certifies claim coverage nor changes validity, settlement or readiness. A diagnostic can still expose genuine harm.
- English comparison
- Complete, careful English
Declared by the submitter; not a certification that the two inputs preserve the same information.
- Reader exposure
- Reader exposure not recorded as a structured label. A visible reference is not training the model’s weights; future Ainglish-trained performance remains unmeasured.
- Condition coverage
- Separate outcomes retained for all 7 declared conditions. An overall average can hide a weak condition. A condition list is not proof that every form or claim in the proposal was tested.
Inspect the declared comparison and reader scope
Comparison label: complete-careful-english-v1
The careful-English arm carries the explicit control phrase (', just this once.' / ', from now on.') in the same fixed final position; the marked arm carries the tag. Frames, contexts and policy clauses are byte-identical across arms.
Exposure label: Not recorded
Reader population: Not recorded
Conditions: once:retry · once:beyond · standing:within · standing:outside · dx:storage-forbidden · dx:audit-required · dx:project-scope
These are the submitter’s declarations, not a certification that the comparison is fair. Bare wording, complete English and visible-reference studies answer different questions; do not pool them by metric name alone.
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.
An external input location was declared, but it is not a safe HTTPS link this view can display. Inspect the retained specification; the missing link does not establish that there were no inputs.
Instrument checks, not language results. Controls deliberately plant a recoverable difference. Check whether answering requires understanding, or merely copying a supplied answer. Passing an answer-copying control does not establish sensitivity to the language distinction.
These are the retained control inputs and keys. They are excluded from study-item totals. The experiment’s reported language score is not a control score.
No readable calibration control pairs are stored inline in this receipt. This does not mean the experiment used none.
Recorded input digest: 7463a0a4fca412475403acad2ab14c191b6199ddc31cbdae20f3b9374f813c40
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.
How to read this receipt
Historical value
This is inactive history. Its reported value is preserved, but it cannot currently support or oppose inclusion.
A reader-panel result does not establish token savings or performance for models outside its declared population.Inactive history
This is inactive history. Its reported value is preserved, but it cannot currently support or oppose inclusion.
Read the public explanation and any corrected successor. Do not replicate this as an active original.One receipt, not the whole decision
No single row ratifies or rejects a proposal. Settlement, every declared metric, deterministic gates and the public ballot remain separate.
This result applies to the declared reader population and exposure conditions. Models outside that population, including future Ainglish-trained models, remain unmeasured.What was tested, and how much?
Test questions measure the language claim. Calibration questions check the instrument; they are not extra evidence for that claim.
- Planned test questions
- 140
- Planned calibration questions
- 12
- Planned test responses
- 140
- Planned calibration responses
- 24
Separate scored test-response counts are not available in this view. Planned counts are not a substitute for completed responses.
Repeated questions and multiple readers do not automatically create independent observations. Use the study’s sampling and uncertainty method, not a pooled response count, to judge precision.
Reported transport: faults 0; truncated responses 0. Missing or conflicting receipts do not mean zero.
Uncertainty and sample
Reported item-bootstrap interval: -15.0366 to 7.4594 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: 140 · Named readers: 1. These are different units; multiple answers to one case are not new cases.
Does the overall result hide differences between conditions?
| Condition | Reported difference | Reported interval | English accuracy | Ainglish accuracy |
|---|---|---|---|---|
once:retry | 12.5 | Not recorded | 87.50% | 100.00% |
once:beyond | 0 | Not recorded | 100.00% | 100.00% |
standing:within | -6.25 | Not recorded | 100.00% | 93.75% |
standing:outside | -20 | Not recorded | 40.00% | 20.00% |
dx:storage-forbidden | 35.71 | Not recorded | 64.29% | 100.00% |
dx:audit-required | 0 | Not recorded | 100.00% | 100.00% |
dx:project-scope | -47.62 | Not recorded | 83.33% | 35.71% |
A missing condition interval is not zero uncertainty. An overall interval cannot substitute for agreement in every load-bearing condition.
Panel
Neff 1 · declared reader count; reader independence is not server-validated
deepseek/deepseek-v4-flash@provider-served
no per-member results declared — divergence structure NOT COMPUTED (aggregate only)
Replication chain
No replications are recorded here. This inactive result is retained for audit, not offered as an active replication target.
Inspect the original manifest — exact, re-runnable specification
These are the committed bytes rendered as readable JSON. Expanding this audit detail does not change the measurement’s current status.
{
"construct": "this-once / from-now-on",
"metric": "comprehension_accuracy_delta",
"seed": 37,
"comparator": {
"kind": "complete-careful-english-v1",
"description": "The careful-English arm carries the explicit control phrase (', just this once.' / ', from now on.') in the same fixed final position; the marked arm carries the tag. Frames, contexts and policy clauses are byte-identical across arms."
},
"items_sha256": "7463a0a4fca412475403acad2ab14c191b6199ddc31cbdae20f3b9374f813c40",
"items_url": "items.json",
"models": [
"deepseek/deepseek-v4-flash@provider-served"
],
"readers": [
{
"name": "deepseek/deepseek-v4-flash",
"provider": "nous-portal-direct",
"model": "deepseek/deepseek-v4-flash",
"precision": "provider-served",
"api": "openai",
"base_url": "https://inference-api.nousresearch.com/v1",
"model_digest": null,
"digest_source": "provider-catalog:openai:/models",
"model_catalog": "openai:/models",
"model_catalog_binding": {
"source": "openai:/models",
"requested_model": "deepseek/deepseek-v4-flash",
"entry_sha256": "sha256:33d0b2cc198d72fd49d6eea961cc70f1f51b9393053ae9e61aa6c8d5052f44f3",
"weight_identity": "provider-opaque"
},
"instrument_preparation": {
"entry_point": "prepare_reader_instruments",
"binding": "provider-catalog:openai:/models"
},
"answer_protocol": "opaque-choice-v1",
"max_tokens": 4096,
"timeout_s": 180,
"temperature": 0,
"seed": "provider-default",
"top_p": "provider-default",
"top_k": "provider-default",
"num_ctx": "provider-default",
"reasoning_effort": "provider-default"
}
],
"instrument_preparation": {
"entry_point": "prepare_reader_instruments",
"binding": [
{
"reader": "deepseek/deepseek-v4-flash@provider-served",
"digest_source": "provider-catalog:openai:/models"
}
]
},
"item_counts": {
"real": 140,
"calibration": 12
},
"interval_kind": "bootstrap_items",
"interval_estimator": {
"kind": "ainglish.panel.bootstrap-items-attestation.v1",
"algorithm": "sha256-counter-modulo-v1",
"draws": 2000,
"sampling_unit": "item",
"quantiles": [
"0.025",
"0.975"
],
"items_index_sha256": "1bea00cf755e5f162e74fd89562839052941c2eb4c9225430971c3550edb11d6"
},
"settlement_strata": [
{
"id": "once:retry",
"weight": 1
},
{
"id": "once:beyond",
"weight": 1
},
{
"id": "standing:within",
"weight": 1
},
{
"id": "standing:outside",
"weight": 1
},
{
"id": "dx:storage-forbidden",
"weight": 1
},
{
"id": "dx:audit-required",
"weight": 1
},
{
"id": "dx:project-scope",
"weight": 1
}
],
"settlement_item_field": "settlement_stratum",
"settlement_rule": "manifest-weighted arms and value; every stratum load-bearing",
"calibration": {
"planted_arm": "ainglish",
"min_gap": 0.125,
"min_recovered": 0.5,
"rule": "headroom-relative-v1",
"ordering": "calibration-first",
"arm_exposure": "both-arms-per-reader-item",
"cells": 24
},
"difficulty": {
"annotated": false
},
"harness": "ainglish-panel/0.2.48",
"transport": {
"deepseek/deepseek-v4-flash@provider-served": {
"max_tokens": 4096,
"timeout_s": 180,
"temperature": 0,
"seed": "provider-default",
"top_p": "provider-default",
"top_k": "provider-default",
"num_ctx": "provider-default",
"reasoning_effort": "provider-default"
}
},
"concurrency": {
"max_in_flight": 1,
"per_reader_max_in_flight": {
"deepseek/deepseek-v4-flash": 1
},
"result_order": "deterministic-plan-order",
"calibration_barrier": true,
"automatic_retries": false
},
"transport_faults": {
"total": 0,
"retried": false,
"per_cell": []
},
"transport_truncations": {
"total": 0,
"per_reader_cell": [],
"by_cell": {
"english": 0,
"ainglish": 0
},
"imbalanced_across_cells": false
},
"protocol": "panel.py counterbalanced real arms + both-arms-per-reader-item planted-effect calibration gate"
}