RISE Humanities Data Benchmark, 0.5.0-pre1

Search Test Runs

 

A test run is a single execution of a benchmark test using a defined model configuration.
Each run represents how a particular large language model (LLM) — such as GPT-4, Claude-3, or Gemini — performed on a given task at a specific time, with specific settings.

A test run includes:

  • Prompt and role definition – what the model was asked to do and from what perspective (e.g. “as a historian”).
  • Model configuration – provider, model version, temperature, and other generation parameters.
  • Results – the model’s actual response and its evaluation (scores such as F1 or accuracy).
  • Usage and cost data – token counts and calculated API costs.
  • Metadata – information like the test date, benchmark name, and person who executed it.

Together, test runs make it possible to compare models, providers, and configurations across benchmarks in a transparent and reproducible way.

Search Results
Show compact results Refine Search New Search
Download JSON Download CSV

Your search for Benchmark 'library_cards__true' with Search Hidden 'False' returned 105 results, showing page 4 of 11.
Result 31 of 105

Test T0440 at 2026-01-24

{'document-type': ['index-card'], 'writing': ['typed', 'printed', 'handwritten'], 'century': [20, 19], 'language': ['de', 'fr', 'en', 'la', 'el', 'fi', 'sv', 'pl'], 'layout': ['index'], 'entry-type': ['bibliographic'], 'task': ['information-extraction']}

Configuration
Providermistral
Modelmistral-small-2506
  
Temperature0.0
DataclassDocument
  
Normalized Score0.00 %
Test timeunknown seconds
Prompt 

Extract the bibliographic information about a historical dissertation from this index card and return it as a structured JSON object with the following exact format:

```json
{
  "type": {
    "type": "Dissertation or thesis" OR "Reference"
  },
  "author": {
    "last_name": "string",
    "first_name": "string"
  },
  "publication": {
    "title": "string",
    "year": integer,
    "place": "string or empty string",
    "pages": "string or empty string",
    "publisher": "string or empty string",
    "format": "string or empty string"
  },
  "library_reference": {
    "shelfmark": "string or empty string",
    "subjects": "string or empty string"
  }
}
```

EXTRACTION RULES:
1. **Card Type**: If a card contains the note "s." on a separate line, it is a "Reference". Otherwise, it is a "Dissertation or thesis".

2. **Author**: Extract last_name and first_name. If only one name is given, put it in last_name and leave first_name empty.

3. **Publication**:
   - title: The main title of the work
   - year: Publication year as integer
   - place: Publication place
   - pages: Page count (remove " S." suffix if present)
   - publisher: Publishing house/institution
   - format: Usually "8°", "8'", or "4°" - single value only

4. **Library Reference**:
   - shelfmark: Often begins with "Diss." or "AT", may be marked with "Standort:"
   - subjects: Subject classifications or keywords

5. **Missing Information**: Use empty string "" for missing text fields, omit year fields entirely if not present.

6. **Ignore**: Disregard any information that doesn't fit into these categories.

Return ONLY the JSON object, no additional text or explanation.

Results 

no valid result

Scoring
Fuzzy Score F1 micro / macro Micro precision/recall Tue/False Positives
n/a 0.00 0.00 0.00 0.00 263 0 0 2415
      Micro Precision Micro Recall Instances TP FP FN
Costs / Pricing
Pricing Date: n/a, n/a. Tokens: 314.5K IT + 40.7K OT = 355.2K TTCost: 0.031$ + 0.012$ = 0.044$
Cite: Hindermann, Marti, Alberto, et al., (2025). RISE-UNIBAS/humanities_data_benchmark, 10.5281/zenodo.16941752
Result 32 of 105

Test T0200 at 2026-01-24

{'document-type': ['index-card'], 'writing': ['typed', 'printed', 'handwritten'], 'century': [20, 19], 'language': ['de', 'fr', 'en', 'la', 'el', 'fi', 'sv', 'pl'], 'layout': ['index'], 'entry-type': ['bibliographic'], 'task': ['information-extraction']}

Configuration
Providergenai
Modelgemini-2.5-flash
  
Temperature0.0
DataclassDocument
  
Normalized Score86.37 %
Test timeunknown seconds
Prompt 

Extract the bibliographic information about a historical dissertation from this index card and return it as a structured JSON object with the following exact format:

```json
{
  "type": {
    "type": "Dissertation or thesis" OR "Reference"
  },
  "author": {
    "last_name": "string",
    "first_name": "string"
  },
  "publication": {
    "title": "string",
    "year": integer,
    "place": "string or empty string",
    "pages": "string or empty string",
    "publisher": "string or empty string",
    "format": "string or empty string"
  },
  "library_reference": {
    "shelfmark": "string or empty string",
    "subjects": "string or empty string"
  }
}
```

EXTRACTION RULES:
1. **Card Type**: If a card contains the note "s." on a separate line, it is a "Reference". Otherwise, it is a "Dissertation or thesis".

2. **Author**: Extract last_name and first_name. If only one name is given, put it in last_name and leave first_name empty.

3. **Publication**:
   - title: The main title of the work
   - year: Publication year as integer
   - place: Publication place
   - pages: Page count (remove " S." suffix if present)
   - publisher: Publishing house/institution
   - format: Usually "8°", "8'", or "4°" - single value only

4. **Library Reference**:
   - shelfmark: Often begins with "Diss." or "AT", may be marked with "Standort:"
   - subjects: Subject classifications or keywords

5. **Missing Information**: Use empty string "" for missing text fields, omit year fields entirely if not present.

6. **Ignore**: Disregard any information that doesn't fit into these categories.

Return ONLY the JSON object, no additional text or explanation.

Results 

no valid result

Scoring
Fuzzy Score F1 micro / macro Micro precision/recall Tue/False Positives
n/a 0.87 0.86 0.85 0.89 263 2149 379 266
      Micro Precision Micro Recall Instances TP FP FN
Costs / Pricing
Pricing Date: n/a, n/a. Tokens: 187.3K IT + 48.1K OT = 235.4K TTCost: 0.056$ + 0.120$ = 0.176$
Cite: Hindermann, Marti, Alberto, et al., (2025). RISE-UNIBAS/humanities_data_benchmark, 10.5281/zenodo.16941752
Result 33 of 105

Test T0430 at 2026-01-24

{'document-type': ['index-card'], 'writing': ['typed', 'printed', 'handwritten'], 'century': [20, 19], 'language': ['de', 'fr', 'en', 'la', 'el', 'fi', 'sv', 'pl'], 'layout': ['index'], 'entry-type': ['bibliographic'], 'task': ['information-extraction']}

Configuration
Providermistral
Modelmagistral-medium-2509
  
Temperature0.0
DataclassDocument
  
Normalized Score0.04 %
Test timeunknown seconds
Prompt 

Extract the bibliographic information about a historical dissertation from this index card and return it as a structured JSON object with the following exact format:

```json
{
  "type": {
    "type": "Dissertation or thesis" OR "Reference"
  },
  "author": {
    "last_name": "string",
    "first_name": "string"
  },
  "publication": {
    "title": "string",
    "year": integer,
    "place": "string or empty string",
    "pages": "string or empty string",
    "publisher": "string or empty string",
    "format": "string or empty string"
  },
  "library_reference": {
    "shelfmark": "string or empty string",
    "subjects": "string or empty string"
  }
}
```

EXTRACTION RULES:
1. **Card Type**: If a card contains the note "s." on a separate line, it is a "Reference". Otherwise, it is a "Dissertation or thesis".

2. **Author**: Extract last_name and first_name. If only one name is given, put it in last_name and leave first_name empty.

3. **Publication**:
   - title: The main title of the work
   - year: Publication year as integer
   - place: Publication place
   - pages: Page count (remove " S." suffix if present)
   - publisher: Publishing house/institution
   - format: Usually "8°", "8'", or "4°" - single value only

4. **Library Reference**:
   - shelfmark: Often begins with "Diss." or "AT", may be marked with "Standort:"
   - subjects: Subject classifications or keywords

5. **Missing Information**: Use empty string "" for missing text fields, omit year fields entirely if not present.

6. **Ignore**: Disregard any information that doesn't fit into these categories.

Return ONLY the JSON object, no additional text or explanation.

Results 

no valid result

Scoring
Fuzzy Score F1 micro / macro Micro precision/recall Tue/False Positives
n/a 0.00 0.00 0.12 0.00 263 1 7 2414
      Micro Precision Micro Recall Instances TP FP FN
Costs / Pricing
Pricing Date: n/a, n/a. Tokens: 314.5K IT + 30.3K OT = 344.8K TTCost: 0.629$ + 0.151$ = 0.780$
Cite: Hindermann, Marti, Alberto, et al., (2025). RISE-UNIBAS/humanities_data_benchmark, 10.5281/zenodo.16941752
Result 34 of 105

Test T0161 at 2026-01-24

{'document-type': ['index-card'], 'writing': ['typed', 'printed', 'handwritten'], 'century': [20, 19], 'language': ['de', 'fr', 'en', 'la', 'el', 'fi', 'sv', 'pl'], 'layout': ['index'], 'entry-type': ['bibliographic'], 'task': ['information-extraction']}

Configuration
Provideropenai
Modelgpt-4.1-mini-2025-04-14
  
Temperature0.0
DataclassDocument
  
Normalized Score46.49 %
Test timeunknown seconds
Prompt 

Extract the bibliographic information about a historical dissertation from this index card and return it as a structured JSON object with the following exact format:

```json
{
  "type": {
    "type": "Dissertation or thesis" OR "Reference"
  },
  "author": {
    "last_name": "string",
    "first_name": "string"
  },
  "publication": {
    "title": "string",
    "year": integer,
    "place": "string or empty string",
    "pages": "string or empty string",
    "publisher": "string or empty string",
    "format": "string or empty string"
  },
  "library_reference": {
    "shelfmark": "string or empty string",
    "subjects": "string or empty string"
  }
}
```

EXTRACTION RULES:
1. **Card Type**: If a card contains the note "s." on a separate line, it is a "Reference". Otherwise, it is a "Dissertation or thesis".

2. **Author**: Extract last_name and first_name. If only one name is given, put it in last_name and leave first_name empty.

3. **Publication**:
   - title: The main title of the work
   - year: Publication year as integer
   - place: Publication place
   - pages: Page count (remove " S." suffix if present)
   - publisher: Publishing house/institution
   - format: Usually "8°", "8'", or "4°" - single value only

4. **Library Reference**:
   - shelfmark: Often begins with "Diss." or "AT", may be marked with "Standort:"
   - subjects: Subject classifications or keywords

5. **Missing Information**: Use empty string "" for missing text fields, omit year fields entirely if not present.

6. **Ignore**: Disregard any information that doesn't fit into these categories.

Return ONLY the JSON object, no additional text or explanation.

Results 

no valid result

Scoring
Fuzzy Score F1 micro / macro Micro precision/recall Tue/False Positives
n/a 0.58 0.46 0.76 0.47 263 1136 355 1279
      Micro Precision Micro Recall Instances TP FP FN
Costs / Pricing
Pricing Date: n/a, n/a. Tokens: 531.9K IT + 72.3K OT = 604.2K TTCost: 0.213$ + 0.116$ = 0.328$
Cite: Hindermann, Marti, Alberto, et al., (2025). RISE-UNIBAS/humanities_data_benchmark, 10.5281/zenodo.16941752
Result 35 of 105

Test T0216 at 2026-01-24

{'document-type': ['index-card'], 'writing': ['typed', 'printed', 'handwritten'], 'century': [20, 19], 'language': ['de', 'fr', 'en', 'la', 'el', 'fi', 'sv', 'pl'], 'layout': ['index'], 'entry-type': ['bibliographic'], 'task': ['information-extraction']}

Configuration
Providergenai
Modelgemini-2.5-flash-lite-preview-09-2025
  
Temperature0.0
DataclassDocument
  
Normalized Score64.85 %
Test timeunknown seconds
Prompt 

Extract the bibliographic information about a historical dissertation from this index card and return it as a structured JSON object with the following exact format:

```json
{
  "type": {
    "type": "Dissertation or thesis" OR "Reference"
  },
  "author": {
    "last_name": "string",
    "first_name": "string"
  },
  "publication": {
    "title": "string",
    "year": integer,
    "place": "string or empty string",
    "pages": "string or empty string",
    "publisher": "string or empty string",
    "format": "string or empty string"
  },
  "library_reference": {
    "shelfmark": "string or empty string",
    "subjects": "string or empty string"
  }
}
```

EXTRACTION RULES:
1. **Card Type**: If a card contains the note "s." on a separate line, it is a "Reference". Otherwise, it is a "Dissertation or thesis".

2. **Author**: Extract last_name and first_name. If only one name is given, put it in last_name and leave first_name empty.

3. **Publication**:
   - title: The main title of the work
   - year: Publication year as integer
   - place: Publication place
   - pages: Page count (remove " S." suffix if present)
   - publisher: Publishing house/institution
   - format: Usually "8°", "8'", or "4°" - single value only

4. **Library Reference**:
   - shelfmark: Often begins with "Diss." or "AT", may be marked with "Standort:"
   - subjects: Subject classifications or keywords

5. **Missing Information**: Use empty string "" for missing text fields, omit year fields entirely if not present.

6. **Ignore**: Disregard any information that doesn't fit into these categories.

Return ONLY the JSON object, no additional text or explanation.

Results 

no valid result

Scoring
Fuzzy Score F1 micro / macro Micro precision/recall Tue/False Positives
n/a 0.66 0.65 0.69 0.63 263 1533 677 882
      Micro Precision Micro Recall Instances TP FP FN
Costs / Pricing
Pricing Date: n/a, n/a. Tokens: 187.3K IT + 503.0K OT = 690.2K TTCost: 0.019$ + 0.201$ = 0.220$
Cite: Hindermann, Marti, Alberto, et al., (2025). RISE-UNIBAS/humanities_data_benchmark, 10.5281/zenodo.16941752
Result 36 of 105

Test T0155 at 2026-01-24

{'document-type': ['index-card'], 'writing': ['typed', 'printed', 'handwritten'], 'century': [20, 19], 'language': ['de', 'fr', 'en', 'la', 'el', 'fi', 'sv', 'pl'], 'layout': ['index'], 'entry-type': ['bibliographic'], 'task': ['information-extraction']}

Configuration
Providergenai
Modelgemini-2.5-pro
  
Temperature0.0
DataclassDocument
  
Normalized Score86.39 %
Test timeunknown seconds
Prompt 

Extract the bibliographic information about a historical dissertation from this index card and return it as a structured JSON object with the following exact format:

```json
{
  "type": {
    "type": "Dissertation or thesis" OR "Reference"
  },
  "author": {
    "last_name": "string",
    "first_name": "string"
  },
  "publication": {
    "title": "string",
    "year": integer,
    "place": "string or empty string",
    "pages": "string or empty string",
    "publisher": "string or empty string",
    "format": "string or empty string"
  },
  "library_reference": {
    "shelfmark": "string or empty string",
    "subjects": "string or empty string"
  }
}
```

EXTRACTION RULES:
1. **Card Type**: If a card contains the note "s." on a separate line, it is a "Reference". Otherwise, it is a "Dissertation or thesis".

2. **Author**: Extract last_name and first_name. If only one name is given, put it in last_name and leave first_name empty.

3. **Publication**:
   - title: The main title of the work
   - year: Publication year as integer
   - place: Publication place
   - pages: Page count (remove " S." suffix if present)
   - publisher: Publishing house/institution
   - format: Usually "8°", "8'", or "4°" - single value only

4. **Library Reference**:
   - shelfmark: Often begins with "Diss." or "AT", may be marked with "Standort:"
   - subjects: Subject classifications or keywords

5. **Missing Information**: Use empty string "" for missing text fields, omit year fields entirely if not present.

6. **Ignore**: Disregard any information that doesn't fit into these categories.

Return ONLY the JSON object, no additional text or explanation.

Results 

no valid result

Scoring
Fuzzy Score F1 micro / macro Micro precision/recall Tue/False Positives
n/a 0.87 0.86 0.85 0.89 263 2147 385 268
      Micro Precision Micro Recall Instances TP FP FN
Costs / Pricing
Pricing Date: n/a, n/a. Tokens: 187.3K IT + 48.9K OT = 236.2K TTCost: 0.234$ + 0.489$ = 0.723$
Cite: Hindermann, Marti, Alberto, et al., (2025). RISE-UNIBAS/humanities_data_benchmark, 10.5281/zenodo.16941752
Result 37 of 105

Test T0152 at 2026-01-24

{'document-type': ['index-card'], 'writing': ['typed', 'printed', 'handwritten'], 'century': [20, 19], 'language': ['de', 'fr', 'en', 'la', 'el', 'fi', 'sv', 'pl'], 'layout': ['index'], 'entry-type': ['bibliographic'], 'task': ['information-extraction']}

Configuration
Providergenai
Modelgemini-2.0-flash-lite
  
Temperature0.0
DataclassDocument
  
Normalized Score81.01 %
Test timeunknown seconds
Prompt 

Extract the bibliographic information about a historical dissertation from this index card and return it as a structured JSON object with the following exact format:

```json
{
  "type": {
    "type": "Dissertation or thesis" OR "Reference"
  },
  "author": {
    "last_name": "string",
    "first_name": "string"
  },
  "publication": {
    "title": "string",
    "year": integer,
    "place": "string or empty string",
    "pages": "string or empty string",
    "publisher": "string or empty string",
    "format": "string or empty string"
  },
  "library_reference": {
    "shelfmark": "string or empty string",
    "subjects": "string or empty string"
  }
}
```

EXTRACTION RULES:
1. **Card Type**: If a card contains the note "s." on a separate line, it is a "Reference". Otherwise, it is a "Dissertation or thesis".

2. **Author**: Extract last_name and first_name. If only one name is given, put it in last_name and leave first_name empty.

3. **Publication**:
   - title: The main title of the work
   - year: Publication year as integer
   - place: Publication place
   - pages: Page count (remove " S." suffix if present)
   - publisher: Publishing house/institution
   - format: Usually "8°", "8'", or "4°" - single value only

4. **Library Reference**:
   - shelfmark: Often begins with "Diss." or "AT", may be marked with "Standort:"
   - subjects: Subject classifications or keywords

5. **Missing Information**: Use empty string "" for missing text fields, omit year fields entirely if not present.

6. **Ignore**: Disregard any information that doesn't fit into these categories.

Return ONLY the JSON object, no additional text or explanation.

Results 

no valid result

Scoring
Fuzzy Score F1 micro / macro Micro precision/recall Tue/False Positives
n/a 0.82 0.81 0.84 0.80 263 1931 369 484
      Micro Precision Micro Recall Instances TP FP FN
Costs / Pricing
Pricing Date: n/a, n/a. Tokens: 676.7K IT + 44.3K OT = 721.0K TTCost: 0.051$ + 0.013$ = 0.064$
Cite: Hindermann, Marti, Alberto, et al., (2025). RISE-UNIBAS/humanities_data_benchmark, 10.5281/zenodo.16941752
Result 38 of 105

Test T0179 at 2026-01-24

{'document-type': ['index-card'], 'writing': ['typed', 'printed', 'handwritten'], 'century': [20, 19], 'language': ['de', 'fr', 'en', 'la', 'el', 'fi', 'sv', 'pl'], 'layout': ['index'], 'entry-type': ['bibliographic'], 'task': ['information-extraction']}

Configuration
Providermistral
Modelmistral-medium-2508
  
Temperature0.0
DataclassDocument
  
Normalized Score0.28 %
Test timeunknown seconds
Prompt 

Extract the bibliographic information about a historical dissertation from this index card and return it as a structured JSON object with the following exact format:

```json
{
  "type": {
    "type": "Dissertation or thesis" OR "Reference"
  },
  "author": {
    "last_name": "string",
    "first_name": "string"
  },
  "publication": {
    "title": "string",
    "year": integer,
    "place": "string or empty string",
    "pages": "string or empty string",
    "publisher": "string or empty string",
    "format": "string or empty string"
  },
  "library_reference": {
    "shelfmark": "string or empty string",
    "subjects": "string or empty string"
  }
}
```

EXTRACTION RULES:
1. **Card Type**: If a card contains the note "s." on a separate line, it is a "Reference". Otherwise, it is a "Dissertation or thesis".

2. **Author**: Extract last_name and first_name. If only one name is given, put it in last_name and leave first_name empty.

3. **Publication**:
   - title: The main title of the work
   - year: Publication year as integer
   - place: Publication place
   - pages: Page count (remove " S." suffix if present)
   - publisher: Publishing house/institution
   - format: Usually "8°", "8'", or "4°" - single value only

4. **Library Reference**:
   - shelfmark: Often begins with "Diss." or "AT", may be marked with "Standort:"
   - subjects: Subject classifications or keywords

5. **Missing Information**: Use empty string "" for missing text fields, omit year fields entirely if not present.

6. **Ignore**: Disregard any information that doesn't fit into these categories.

Return ONLY the JSON object, no additional text or explanation.

Results 

no valid result

Scoring
Fuzzy Score F1 micro / macro Micro precision/recall Tue/False Positives
n/a 0.01 0.00 0.09 0.00 263 7 67 2408
      Micro Precision Micro Recall Instances TP FP FN
Costs / Pricing
Pricing Date: n/a, n/a. Tokens: 314.5K IT + 41.2K OT = 355.7K TTCost: 0.126$ + 0.082$ = 0.208$
Cite: Hindermann, Marti, Alberto, et al., (2025). RISE-UNIBAS/humanities_data_benchmark, 10.5281/zenodo.16941752
Result 39 of 105

Test T0148 at 2026-01-24

{'document-type': ['index-card'], 'writing': ['typed', 'printed', 'handwritten'], 'century': [20, 19], 'language': ['de', 'fr', 'en', 'la', 'el', 'fi', 'sv', 'pl'], 'layout': ['index'], 'entry-type': ['bibliographic'], 'task': ['information-extraction']}

Configuration
Provideranthropic
Modelclaude-sonnet-4-20250514
  
Temperature0.0
DataclassDocument
  
Normalized Score84.02 %
Test timeunknown seconds
Prompt 

Extract the bibliographic information about a historical dissertation from this index card and return it as a structured JSON object with the following exact format:

```json
{
  "type": {
    "type": "Dissertation or thesis" OR "Reference"
  },
  "author": {
    "last_name": "string",
    "first_name": "string"
  },
  "publication": {
    "title": "string",
    "year": integer,
    "place": "string or empty string",
    "pages": "string or empty string",
    "publisher": "string or empty string",
    "format": "string or empty string"
  },
  "library_reference": {
    "shelfmark": "string or empty string",
    "subjects": "string or empty string"
  }
}
```

EXTRACTION RULES:
1. **Card Type**: If a card contains the note "s." on a separate line, it is a "Reference". Otherwise, it is a "Dissertation or thesis".

2. **Author**: Extract last_name and first_name. If only one name is given, put it in last_name and leave first_name empty.

3. **Publication**:
   - title: The main title of the work
   - year: Publication year as integer
   - place: Publication place
   - pages: Page count (remove " S." suffix if present)
   - publisher: Publishing house/institution
   - format: Usually "8°", "8'", or "4°" - single value only

4. **Library Reference**:
   - shelfmark: Often begins with "Diss." or "AT", may be marked with "Standort:"
   - subjects: Subject classifications or keywords

5. **Missing Information**: Use empty string "" for missing text fields, omit year fields entirely if not present.

6. **Ignore**: Disregard any information that doesn't fit into these categories.

Return ONLY the JSON object, no additional text or explanation.

Results 

no valid result

Scoring
Fuzzy Score F1 micro / macro Micro precision/recall Tue/False Positives
n/a 0.85 0.84 0.84 0.86 263 2088 408 327
      Micro Precision Micro Recall Instances TP FP FN
Costs / Pricing
Pricing Date: n/a, n/a. Tokens: 634.2K IT + 54.5K OT = 688.6K TTCost: 1.903$ + 0.817$ = 2.720$
Cite: Hindermann, Marti, Alberto, et al., (2025). RISE-UNIBAS/humanities_data_benchmark, 10.5281/zenodo.16941752
Result 40 of 105

Test T0160 at 2026-01-24

{'document-type': ['index-card'], 'writing': ['typed', 'printed', 'handwritten'], 'century': [20, 19], 'language': ['de', 'fr', 'en', 'la', 'el', 'fi', 'sv', 'pl'], 'layout': ['index'], 'entry-type': ['bibliographic'], 'task': ['information-extraction']}

Configuration
Provideropenai
Modelgpt-4.1-2025-04-14
  
Temperature0.0
DataclassDocument
  
Normalized Score84.20 %
Test timeunknown seconds
Prompt 

Extract the bibliographic information about a historical dissertation from this index card and return it as a structured JSON object with the following exact format:

```json
{
  "type": {
    "type": "Dissertation or thesis" OR "Reference"
  },
  "author": {
    "last_name": "string",
    "first_name": "string"
  },
  "publication": {
    "title": "string",
    "year": integer,
    "place": "string or empty string",
    "pages": "string or empty string",
    "publisher": "string or empty string",
    "format": "string or empty string"
  },
  "library_reference": {
    "shelfmark": "string or empty string",
    "subjects": "string or empty string"
  }
}
```

EXTRACTION RULES:
1. **Card Type**: If a card contains the note "s." on a separate line, it is a "Reference". Otherwise, it is a "Dissertation or thesis".

2. **Author**: Extract last_name and first_name. If only one name is given, put it in last_name and leave first_name empty.

3. **Publication**:
   - title: The main title of the work
   - year: Publication year as integer
   - place: Publication place
   - pages: Page count (remove " S." suffix if present)
   - publisher: Publishing house/institution
   - format: Usually "8°", "8'", or "4°" - single value only

4. **Library Reference**:
   - shelfmark: Often begins with "Diss." or "AT", may be marked with "Standort:"
   - subjects: Subject classifications or keywords

5. **Missing Information**: Use empty string "" for missing text fields, omit year fields entirely if not present.

6. **Ignore**: Disregard any information that doesn't fit into these categories.

Return ONLY the JSON object, no additional text or explanation.

Results 

no valid result

Scoring
Fuzzy Score F1 micro / macro Micro precision/recall Tue/False Positives
n/a 0.85 0.84 0.85 0.84 263 2033 356 382
      Micro Precision Micro Recall Instances TP FP FN
Costs / Pricing
Pricing Date: n/a, n/a. Tokens: 457.8K IT + 361.0K OT = 818.9K TTCost: 0.916$ + 2.888$ = 3.804$
Cite: Hindermann, Marti, Alberto, et al., (2025). RISE-UNIBAS/humanities_data_benchmark, 10.5281/zenodo.16941752