qwen3-4b-yaml-sft-with-template/chat_template.jinja

{%- if messages[0].role == 'system' %}
    {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
{%- endif %}
{%- for message in messages %}
    {%- if message.role == "user" or (message.role == "system" and not loop.first) %}
        {%- set content = message.content if message.content is string else '' %}
        {%- set suffix = '' %}
        {%- set fl = content.split('\n')[0].lower() %}
        {%- set formats = ['toml', 'xml', 'yaml', 'json', 'csv'] %}
        {%- set ns = namespace(target='') %}
        {%- for f in formats %}
            {%- if f in fl %}
                {%- if ns.target == '' or fl.rfind(f) > fl.rfind(ns.target) %}
                    {%- set ns.target = f %}
                {%- endif %}
            {%- endif %}
        {%- endfor %}
        {%- if ns.target == 'toml' %}
            {%- set suffix = '\n\nOutput ONLY raw TOML. Use [table] for nested objects and [[array]] for arrays. Do not use inline tables.' %}
        {%- elif ns.target == 'xml' %}
            {%- set suffix = '\n\nOutput ONLY raw XML. Escape & as &amp;, < as &lt;. No XML declaration. Follow these strict rules:

1. ROOT ELEMENT DETERMINATION:
   - For JSON input: Use the FIRST KEY from the input object as root name
   - For data generation queries: Infer based on expected path structure. If paths start with \'assessment_id.*\', root must be \'<assessment_id>\', not the entity name
   - All XML content must be wrapped within this single root element

2. FIELD ORDER & HIERARCHY:
   - When generating data, follow required path structure (e.g., \'assessment_id.assessment_id\' requires:
     <assessment_id>
       <assessment_id>UUID</assessment_id>
       ...other nested fields...
     </assessment_id>)
   - Maintain exact field order required for path consistency
   - Arrays must be nested within their parent elements with _item suffix

3. XML SYNTAX REQUIREMENTS:
   - Must be parseable XML with ONE root element
   - Properly escape special characters
   - Match expected schema\'s path structure exactly
   - All opening tags must have closing tags

4. DATA TYPE HANDLING:
   - Maintain ISO 8601 timestamps (YYYY-MM-DDTHH:MM:SS.ssssss)
   - Preserve UUID formats (e.g., 5ceff241-2713-4bf4-a7b2-a4b795b4a1a9)
   - Keep DOI identifiers intact (e.g., 10.8889/HDtI)

5. SPECIFIC STRUCTURE REQUIREMENTS:
   - Academic assessments must have:
     assessment_id.assessment_id
     assessment_id.student_id
     assessment_id.course_id
     assessment_id.type
     assessment_id.date
   - Arrays must be properly nested (no standalone array containers)
   - Academic structures must maintain:
     - Assignments with name/due/weight
     - Textbooks with title/author/isbn
     - Authors arrays with name/affiliation/email

6. VALIDATION CHECKS:
   - Verify path format matches exactly:
     \'{root_element}.{field_name}\'
     \'{root_element}.{array_field}[index].{subfield}\'
   - Ensure all expected fields are present in correct hierarchy
   - Confirm single root element contains all content
   - Check proper array element formatting with _item suffix' %}
        {%- elif ns.target == 'yaml' %}
            {%- set suffix = '\n\nConvert the input data format (XML, CSV, or JSON) to valid, semantically equivalent YAML following these rules:

1. SYNTAX & FORMATTING
- Use 2-space indentation levels for nesting
- Represent lists with hyphens (-) followed by single space
- Quote string values containing special characters (e.g., "-120", "0-18km")
- Maintain consistent spacing around colons (key: value)
- NO trailing spaces or markdown formatting
- Ensure valid YAML syntax (test with parser before output)

2. STRUCTURE PRESERVATION
- XML Conversion:
  - Map XML tags to nested keys
  - Convert multiple elements with same tag to list of maps DIRECTLY UNDER PARENT KEY, WITHOUT WRAPPING IN TAG NAME:
    ✓ Correct: <planets><planet>Mercury</planet></planets> → planets: ["Mercury"]
    ✓ Correct: <star_systems><star_system><name>Alpha</name></star_system> → star_systems: [ { name: "Alpha" } ]
    ✗ Incorrect: star_systems: [ { star_system: { name: "Alpha" } } ]
  - Preserve element hierarchy: parent-child relationships must create direct key-value pairs
  - Convert XML attributes to keys with \'@\' prefix (e.g., <moon name="Phobos"/> → { "@name": "Phobos" })
- CSV Conversion:
  - Map dot-notation headers to nested structures (system.star_name → system: { star_name: ... })
  - Create object arrays for underscore-prefixed columns (moons_diameter_km → moons: [ { diameter_km: ... } ])
  - Maintain field relationships in grouped columns (moons_diameter_km and moons_has_rings → moons: [ { diameter_km, has_rings } ])
  - Preserve row positions when splitting multi-value fields (Phobos;Deimos and 12;6 → first moon: name=Phobos, diameter=12)
- JSON Conversion:
  - Directly map objects to YAML maps
  - Maintain array order and indexing
  - Preserve null values from source

3. SPECIAL CASE HANDLING
- Empty elements: Represent as null (undefined) or empty string (explicitly empty) based on source
- CSV multi-value fields: 
  - Convert semicolon-separated values to arrays (Oxygen;Nitrogen → ["Oxygen", "Nitrogen"])
  - Align split values by position for grouped fields (moons field1;field2 → moons[*].field)
- Numeric values: Keep unquoted, preserve type (int, float) from source
- Boolean values: Convert to lowercase YAML booleans (true/false)
- String numbers: Quote values starting with "-" or containing "-...km" patterns

4. PATH PRESERVATION & VALIDATION
- Ensure all source paths exist in output using dot-notation:
  - XML: galaxy.clusters[0].name, galaxy.clusters[0].stars[0].type
  - CSV: atmosphere.composition.*, surface.topography, moons.*.moon_name
- Confirm nested structures match source depth exactly
- Validate list items maintain field relationships (moons.*.has_rings)
- Verify key names match source paths including array element properties
- Test YAML validity with parser before output

5. XML SPECIFIC NESTING RULES
- For elements containing mixed content or simple text:
  <planet><![CDATA[Earth;Mars]]></planet> → planet: "Earth;Mars"
- For elements with child elements:
  <planet>
    <name>Proxima b</name>
    <radius>0.12</radius>
  </planet> → planet:
    name: Proxima b
    radius: 0.12
- For repeated elements (critical rule):
  <star_systems>
    <star_system>  <!-- THIS TAG NAME MUST BE ELIDED IN YAML -->
      <name>Alpha Centauri</name>
      <planets>
        <planet>  <!-- ELEMENT TO ARRAY CONVERSION -->
          <name>Proxima b</name>
        </planet>
      </planets>
    </star_system>
  </star_systems> → star_systems:
    - name: Alpha Centauri
      planets:
        - name: Proxima b

6. CSV SPECIFIC MAPPINGS
- Top-level columns (no dot/underscore prefix) go at root level (planet_name: ...)
- Convert dot notation headers to nested maps (system.star_name → system: { star_name: ... })
- Map underscore-prefixed columns to object arrays:
  - moons_diameter_km → moons: [ { diameter_km: ... } ]
  - moons_has_rings → moons: [ { has_rings: ... } ]
- Preserve field relationships in grouped columns:
  - moons, moons_diameter_km → moons: [ { moon_name, diameter_km } ]
  - moons_diameter_km, moons_has_rings → moons: [ { diameter_km, has_rings } ]

OUTPUT ONLY THE RAW YAML WITH NO EXPLANATIONS. MAINTAIN 100% STRUCTURAL PARITY WITH SOURCE DATA.' %}
        {%- endif %}
        {{- '<|im_start|>' + message.role + '\n' + content + suffix + '<|im_end|>' + '\n' }}
    {%- elif message.role == "assistant" %}
        {%- set content = message.content if message.content is string else '' %}
        {{- '<|im_start|>assistant\n' + content + '<|im_end|>\n' }}
    {%- endif %}
{%- endfor %}
{%- if add_generation_prompt %}
    {{- '<|im_start|>assistant\n' }}
{%- endif %}