Run
│ name: "Fact Extraction"
│ tags: ['model:mistral-small']
│ metrics:
│   conciseness: 0.890625
│   recall: 0.5
│ Records
│   Record 1/3
│   │ input: {'text': "Kimberly 'Kim' Wong is widely regarded as one of the most talented wildlife photographers of her generation. Her evocative and deeply personal style has resonated with audiences around the world, earning her a passionate following among nature enthusiasts. Over the course of her career, she has documented over 300 species across multiple continents.", 'facts': ['wildlife photographer', '...
│   │ metrics:
│   │   conciseness: 1.0
│   │   recall: 0.5
│   │ Generations
│   │   Generation 1/1
│   │   │ output: ['Kim Wong documented 300+ species']
│   │   │ scores:
│   │   │   conciseness: 1.0
│   │   │   │ [1]: 1.0
│   │   │   recall: 0.5
│   │   │   │ [1]: {'score': 0.5, 'comment': "Only one of the two ground truth facts is captured in the predicted facts. The fact 'wildlife photographer' is missing."}
│   │   │   │ [2]: {'score': 1.0, 'comment': "All ground truth facts are present in the predicted facts. The predicted fact 'Kim Wong documented 300+ species' captures both the ground truth facts 'wildlife photographer' (implied by documenting species) and 'documented 300 species' (with the '+' indicating the exact number is at least 300). No facts are missing."}
│   │   │   │ [3]: {'score': 0.3333333333333333, 'comment': "The ground truth fact 'wildlife photographer' is missing from the predicted facts."}
│   Record 2/3
│   │ input: {'text': 'Avery Johnson has always been fascinated by the stories hidden in old paper and fading ink. His quiet dedication to preserving the past is evident in everything he does, from the careful way he handles fragile manuscripts to the enthusiasm he brings to public talks about archival work. Last year, he digitized over 12,000 documents while working at the Boston Public Library.', 'facts': ['...
│   │ metrics:
│   │   conciseness: 1.0
│   │   recall: 0.5
│   │ Generations
│   │   Generation 1/1
│   │   │ output: ['Avery digitized 12,000 documents']
│   │   │ scores:
│   │   │   conciseness: 1.0
│   │   │   │ [1]: 1.0
│   │   │   recall: 0.5
│   │   │   │ [1]: {'score': 0.5, 'comment': "The predicted facts only capture one out of two ground truth facts. The fact 'works at Boston Public Library' is missing."}
│   │   │   │ [2]: {'score': 0.5, 'comment': "The predicted facts only capture one out of two ground truth facts. The fact 'works at Boston Public Library' is missing."}
│   │   │   │ [3]: {'score': 0.5, 'comment': "The predicted facts only capture one out of two ground truth facts. The fact 'works at Boston Public Library' is missing."}
│   Record 3/3
│   │ input: {'text': "Taylor Brown's empathy and tireless advocacy have made her a beloved figure in her community. Friends and colleagues describe her as someone who never stops looking for ways to help others, especially young people facing difficult circumstances. Her nonprofit has placed over 400 young people in transitional housing since 2019.", 'facts': ['has a nonprofit since 2019', '400 people housed'...
│   │ metrics:
│   │   conciseness: 0.671875
│   │   recall: 1.0
│   │ Generations
│   │   Generation 1/1
│   │   │ output: ["Taylor Brown's nonprofit placed 400+ youth in housing since 2019"]
│   │   │ scores:
│   │   │   conciseness: 0.671875
│   │   │   │ [1]: 0.671875
│   │   │   recall: 1.0
│   │   │   │ [1]: {'score': 1.0, 'comment': "No ground truth facts are missing. The predicted fact captures both the ground truth facts: 'has a nonprofit since 2019' and '400 people housed'."}
│   │   │   │ [2]: {'score': 1.0, 'comment': 'Both ground truth facts are captured in the predicted facts. The predicted facts mention a nonprofit since 2019 and 400 people housed, which matches the ground truth.'}
│   │   │   │ [3]: {'score': 1.0, 'comment': 'All ground truth facts are captured in the predicted facts. The predicted facts mention a nonprofit since 2019 and 400 people housed, which matches the ground truth.'}
