Minto

qwen3.6-fp8 на кейсе «Набор заголовков»: с Minto лучше по обеим мерам

Структура 5 → 6 из 8 +1 Качество 6 → 7 из 10 +1

Судья об ответе с Minto

Has complete apparatus and identifies the main answer/count/order defects, but its SCQ fix fabricates the absent situation and complication.

Режим audit · кейс 07-headings-audit · прогон v1.7.0-final

Структура из 8, по 2 балла на ось
Без Minto 5С Minto 6Изменение
Вершина
Состав ключевой линии
Уровни
Порядок и род+1
Качество из 10, по 2 балла на ось
Без Minto 6С Minto 7Изменение
Вершина
Однородные группы
Объяснимый порядок+1
MECE−1
Видимая соразмерная структура+1

Ответы моделей и слова судьи приведены в оригинале, по-английски.

Без Minto С Minto

Без Minto

контрольный промпт · слепая метка out-16

Жёстких провалов нет

Clearly identifies the purpose statement, mixed levels, and missing ordering framework, but adds unsupported demands for scope and boundary content while omitting several gold defects.

Audit Findings

  • Violates “Answer-First” Principle: The opening leads with a descriptive purpose statement (“This memorandum describes…”) rather than stating the main conclusion, recommendation, or direct answer to the reader’s implicit question.
  • Missing Logical Sequencing Framework: The six headings are presented as a flat list without indicating the underlying logic governing their order (e.g., chronological, structural, causal, or degree-based). MPP requires that supporting ideas follow a single, explicit reasoning path.
  • No Evidence of Top-Down Grouping: The setout shows no hierarchical clustering or subordination. MPP mandates that related concepts be grouped under higher-level themes before being enumerated.
  • Incomplete Contextual Framing: The opening lacks a clear statement of scope, boundary conditions, or the specific business question/decision being addressed, which weakens the necessary setup for an answer-first structure.
  • Inconsistent Conceptual Level Across Headings: The phrasing mixes methodological elements (“Principles,” “Definition,” “Organization”) with outcome-oriented elements (“Benefits and results”) and prerequisites (“Conditions”). This suggests a lack of mutually exclusive, collectively exhaustive categorization at the outline stage, violating MPP’s grouping discipline.

С Minto

промпт с навыком · слепая метка out-09
  • invented_facts

Has complete apparatus and identifies the main answer/count/order defects, but its SCQ fix fabricates the absent situation and complication.

▲ top / answer | ● first-level group | ○ supporting detail | ⚠ violation | ↑ buried conclusion | ⇄ wrong order | ⊗ overlap | ⊕ gap | ✂ cut/appended

[✂][⚠] This memorandum describes how project teams identify and deliver significant
                profit improvements. [✂] It is organized in six sections:
                
                    [●][⊗] Background
                    [●][⊗] Principles of the project-team approach
                    [●][⊗] Definition of project work
                    [●][⊗] Organization of the program
                    [●][⊗] Distinctive benefits and results
                    [●][⊗] Conditions required for success
location → violation → fix
Opening sentence → Topic label instead of answer → Replace with a single-sentence answer to the reader’s decision question
Entire intro → Missing SCQ story → Add Situation (current profit baseline) and Complication (blocker/change) leading to the Question
Section list → Mixed kinds & unexplainable order → Collapse into 3–4 same-kind branches (e.g., reasons, steps, criteria) with explicit ordering
Structure → Table-of-contents format → Restructure as argument-first hierarchy with clear key lines under each branch
Scope → Six items exceed working-memory limit → Consolidate to three or four core points; move remaining details to supports or appendix

Score

  • Top: 0/2
  • Same-kind groups: 0/2
  • Order: 0/2
  • MECE: 0/2
  • Display: 0/2
    Band: 0–3 not structured yet

Three highest-impact fixes

  1. Replace the opening description with a one-sentence answer that directly resolves the reader’s profit-improvement question.
  2. Draft a concise SCQ intro that states what the reader already accepts and what specifically forces a choice now.
  3. Group the six report sections into three or four same-kind branches ordered by time, structure, or ranking, and promote the real recommendation to the top.

Как устроен тест

qwen3.6-fp8 получила одну и ту же задачу дважды. Контрольный промпт — одна вводная строка и задача. Промпт с Minto — та же задача, перед которой стоит полный текст навыка. Судья читал оба ответа среди 16 ответов под слепыми метками и не знал, в каком из них был навык.

Промпт без Minto 998 Б 
You are working on a business document using the Minto Pyramid Principle.

<!-- Faithful paraphrase of Minto (2010), Chapter 4, the six-heading
     project-team example. Original wording for this fixture. -->

# Fixture 07: headings audit

**Mode:** `audit`
**Language:** `en`

## Context

Only the opening and setout below are available. The body of the memorandum is
not available and must not be inferred.

## Before

```text
This memorandum describes how project teams identify and deliver significant
profit improvements. It is organized in six sections:

    Background
    Principles of the project-team approach
    Definition of project work
    Organization of the program
    Distinctive benefits and results
    Conditions required for success
```

## Task

Audit the visible opening against the Minto Pyramid Principle. Report what is
wrong without rewriting the document or inventing facts about the missing body.

Return only the result, with no explanation of how you produced it.
Промпт с Minto пять частей по порядку · 28 571 Б
  1. Below is a skill written as an instruction. Read it in full and apply it to the task at the end.
  2. ===== SKILL.md ===== SKILL.md на f4813ce 17 235 Б
  3. ===== references/rules.md ===== references/rules.md на f4813ce 6 464 Б
  4. ===== references/templates.md ===== references/templates.md на f4813ce 3 742 Б
  5. ===== TASK ===== вход кейса целиком
  6. Return only the result, with no explanation of how you produced it.
Вход кейса before.md

<!-- Faithful paraphrase of Minto (2010), Chapter 4, the six-heading
project-team example. Original wording for this fixture. -->

Fixture 07: headings audit

Mode: audit
Language: en

Context

Only the opening and setout below are available. The body of the memorandum is
not available and must not be inferred.

Before
This memorandum describes how project teams identify and deliver significant
                profit improvements. It is organized in six sections:
                
                    Background
                    Principles of the project-team approach
                    Definition of project work
                    Organization of the program
                    Distinctive benefits and results
                    Conditions required for success
Task

Audit the visible opening against the Minto Pyramid Principle. Report what is
wrong without rewriting the document or inventing facts about the missing body.

Эталон судьи gold.md

Gold 07: headings audit

Source structure: Minto (2010), Chapter 4.

Visible defects
DefectSeverity
The setout contains topics, not ideas, so it conveys no message or key lineCritical
The first sentence describes a subject instead of answering a reader questionCritical
Background cannot sit at the same abstraction level as the other pointsMedium
Benefits/results likely belong under principles; success conditions likely belong under program organizationMedium
Six first-level topics exceed the skill's normal display limitMedium
The topics do not form a visible inductive or deductive argumentMedium

The output must stay within the visible evidence. Claims about missing data,
metrics, proof, vertical support inside sections, or exact content of the body
are false positives. Invented page numbers, line numbers, or quotations are
hard failures.

Судья gpt-5.6-sol, effort high · навык с коммита f4813ce · прогон v1.7.0-final