{"_id": {"$oid": "69d4eb12b877594bc57188dd"}, "test_id": "T0180", "benchmark": "library_cards", "date": "2025-09-26", "tags": {"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"]}, "contributors": ["gabriel_muller", "maximilian_hindermann"], "hidden": false, "config": {"provider": "mistral", "model": "mistral-medium-2505", "dataclass": "Document", "temperature": 0.0, "role_description": "You are a historian with keyword knowledge", "prompt_file": "prompt.txt", "legacy_test": false}, "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:\n\n```json\n{\n  \"type\": {\n    \"type\": \"Dissertation or thesis\" OR \"Reference\"\n  },\n  \"author\": {\n    \"last_name\": \"string\",\n    \"first_name\": \"string\"\n  },\n  \"publication\": {\n    \"title\": \"string\",\n    \"year\": integer,\n    \"place\": \"string or empty string\",\n    \"pages\": \"string or empty string\",\n    \"publisher\": \"string or empty string\",\n    \"format\": \"string or empty string\"\n  },\n  \"library_reference\": {\n    \"shelfmark\": \"string or empty string\",\n    \"subjects\": \"string or empty string\"\n  }\n}\n```\n\nEXTRACTION RULES:\n1. **Card Type**: If a card contains the note \"s.\" on a separate line, it is a \"Reference\". Otherwise, it is a \"Dissertation or thesis\".\n\n2. **Author**: Extract last_name and first_name. If only one name is given, put it in last_name and leave first_name empty.\n\n3. **Publication**:\n   - title: The main title of the work\n   - year: Publication year as integer\n   - place: Publication place\n   - pages: Page count (remove \" S.\" suffix if present)\n   - publisher: Publishing house/institution\n   - format: Usually \"8\u00b0\", \"8'\", or \"4\u00b0\" - single value only\n\n4. **Library Reference**:\n   - shelfmark: Often begins with \"Diss.\" or \"AT\", may be marked with \"Standort:\"\n   - subjects: Subject classifications or keywords\n\n5. **Missing Information**: Use empty string \"\" for missing text fields, omit year fields entirely if not present.\n\n6. **Ignore**: Disregard any information that doesn't fit into these categories.\n\nReturn ONLY the JSON object, no additional text or explanation.", "results": {"provider": "mistral", "model": "mistral-medium-2505", "duration": 3.934082269668579, "timestamp": "2025-09-26T09:42:28.181306", "text": "{\n  \"type\": {\n    \"type\": \"Dissertation or thesis\"\n  },\n  \"author\": {\n    \"last_name\": \"Wolff\",\n    \"first_name\": \"Dorothee\"\n  },\n  \"publication\": {\n    \"title\": \"Der Phosphoglukomutase-Isoenzympolymorphismus in der unterfr\u00e4nkischen Bev\u00f6lkerung. Stichprobe und Vergleich mit den ABO- und Rhesus-Systemen bei 1000 nicht verwandten Probanden\",\n    \"year\": 1971,\n    \"place\": \"W\u00fcrzburg\",\n    \"pages\": \"40\",\n    \"publisher\": \"\",\n    \"format\": \"8\u00b0\"\n  },\n  \"library_reference\": {\n    \"shelfmark\": \"Diss.1972,2843\",\n    \"subjects\": \"\"\n  }\n}", "finish_reason": "stop"}, "scoring": {"f1_micro": 0.7782672540381792, "f1_macro": 0.7755133079847909, "micro_precision": 0.7886904761904762, "micro_recall": 0.7681159420289855, "total_instances": 263, "total_tp": 1855, "total_fp": 497, "total_fn": 560}, "normalized_score": 77.55133079847909}