281 lines
11 KiB
JSON
281 lines
11 KiB
JSON
{
|
|
"validation_rate": 48.148148148148145,
|
|
"correct": 13,
|
|
"total": 27,
|
|
"thresholds": {
|
|
"high": 0.7,
|
|
"low": 0.4
|
|
},
|
|
"results": [
|
|
{
|
|
"name": "Claude: Lubbock on everything album lyrics",
|
|
"subsample": "a",
|
|
"bucket": "high",
|
|
"frame_relationships": "The Songwriting frame is the primary focus of this content, with the Storytelling and Literary Analysis frames providing context and analysis for understanding Terry Allen's writing style.",
|
|
"coverage_score": 1.0,
|
|
"predicted_tier": "baseline",
|
|
"actual_best": "baseline",
|
|
"match": true
|
|
},
|
|
{
|
|
"name": "ChatGPT: Tulsa Concept Album Guide",
|
|
"subsample": "a",
|
|
"bucket": "high",
|
|
"frame_relationships": "The Personal_Narrative and Place_Exploration frames are intertwined, with the album being a personal account of the writer's experiences in Tulsa. The Autobiography frame is used to disguise the concept album as a traditional autobiography.",
|
|
"coverage_score": 1.0,
|
|
"predicted_tier": "baseline",
|
|
"actual_best": "taxfree",
|
|
"match": false
|
|
},
|
|
{
|
|
"name": "ChatGPT: Rhino 3D object flow",
|
|
"subsample": "a",
|
|
"bucket": "high",
|
|
"frame_relationships": "The philosophical concepts provide a framework for understanding and describing digital fabrication processes.",
|
|
"coverage_score": 1.0,
|
|
"predicted_tier": "baseline",
|
|
"actual_best": "baseline",
|
|
"match": true
|
|
},
|
|
{
|
|
"name": "Claude: SUNY faculty conflict of interest policies",
|
|
"subsample": "a",
|
|
"bucket": "mid",
|
|
"frame_relationships": "The document primarily discusses the SUNY conflict of interest policy for faculty members, focusing on outside work and consulting. The system-level policy and specific campus handbooks are provided as resources.",
|
|
"coverage_score": 1.0,
|
|
"predicted_tier": "baseline",
|
|
"actual_best": "baseline",
|
|
"match": true
|
|
},
|
|
{
|
|
"name": "Claude: Interview presentation research and preparation",
|
|
"subsample": "a",
|
|
"bucket": "mid",
|
|
"frame_relationships": "The Digital-Physical Interface frame centers the workflow between digital models and physical objects, while Emerging Technologies highlights the specific tools (AI, XR) used in this workflow.",
|
|
"coverage_score": 1.0,
|
|
"predicted_tier": "baseline",
|
|
"actual_best": "baseline",
|
|
"match": true
|
|
},
|
|
{
|
|
"name": "ChatGPT: Respect Individual Interests for Christmas",
|
|
"subsample": "a",
|
|
"bucket": "low",
|
|
"frame_relationships": "The frames of gift giving and interests/preferences are intertwined as the document discusses choosing gifts based on a recipient's unique hobbies and preferences during the holiday season.",
|
|
"coverage_score": 1.0,
|
|
"predicted_tier": "baseline",
|
|
"actual_best": "baseline",
|
|
"match": true
|
|
},
|
|
{
|
|
"name": "University of North Texas Cover letter.pdf",
|
|
"subsample": "a",
|
|
"bucket": "document",
|
|
"frame_relationships": "The 'Education' and 'Job Application' frames are the primary drivers of the document, with 'Artwork Creation' providing contextual details. The 'Academia' frame serves as a broader context for all other frames.",
|
|
"coverage_score": 1.0,
|
|
"predicted_tier": "baseline",
|
|
"actual_best": "taxfree",
|
|
"match": false
|
|
},
|
|
{
|
|
"name": "Claude: Finding ideal rural housing near University of Utah",
|
|
"subsample": "a",
|
|
"bucket": "high",
|
|
"frame_relationships": "The geographical location and community characteristics inform the housing and transportation options in each area.",
|
|
"coverage_score": 1.0,
|
|
"predicted_tier": "baseline",
|
|
"actual_best": "taxfree",
|
|
"match": false
|
|
},
|
|
{
|
|
"name": "ChatGPT: SEC coaches with OSU ties",
|
|
"subsample": "a",
|
|
"bucket": "high",
|
|
"frame_relationships": "",
|
|
"coverage_score": 0.0,
|
|
"predicted_tier": "taxfree",
|
|
"actual_best": "baseline",
|
|
"match": false
|
|
},
|
|
{
|
|
"name": "Claude: Bonding ASA 3D printed parts",
|
|
"subsample": "a",
|
|
"bucket": "mid",
|
|
"frame_relationships": "The frames 'Bonding' and 'Material_Selection' are active, with the latter providing context for the former by discussing suitable adhesives for bonding 3D printed parts.",
|
|
"coverage_score": 1.0,
|
|
"predicted_tier": "baseline",
|
|
"actual_best": "taxfree",
|
|
"match": false
|
|
},
|
|
{
|
|
"name": "ChatGPT: Title: User request summary.",
|
|
"subsample": "a",
|
|
"bucket": "low",
|
|
"frame_relationships": "The 'Event' frame is the main context, within which the 'Speech', 'Hosting', and 'Communication' frames are active.",
|
|
"coverage_score": 1.0,
|
|
"predicted_tier": "baseline",
|
|
"actual_best": "taxfree",
|
|
"match": false
|
|
},
|
|
{
|
|
"name": "ChatGPT: Scholarship Recommendation Letter Tips",
|
|
"subsample": "a",
|
|
"bucket": "low",
|
|
"frame_relationships": "The frames interact to provide a comprehensive assessment of the candidate's qualifications for a scholarship, focusing on both their academic and personal qualities.",
|
|
"coverage_score": 1.0,
|
|
"predicted_tier": "baseline",
|
|
"actual_best": "baseline",
|
|
"match": true
|
|
},
|
|
{
|
|
"name": "ChatGPT: Job application comparison",
|
|
"subsample": "b",
|
|
"bucket": "high",
|
|
"frame_relationships": "The frames interact by comparing the applicant's qualifications to the job requirements, peer deans' profiles, and the mission of the School of the Arts at Western.",
|
|
"coverage_score": 1.0,
|
|
"predicted_tier": "baseline",
|
|
"actual_best": "standard",
|
|
"match": false
|
|
},
|
|
{
|
|
"name": "ChatGPT: External review for tenure",
|
|
"subsample": "b",
|
|
"bucket": "high",
|
|
"frame_relationships": "The Tenure Review frame is the central and overarching context in which the other frames (Academic Evaluation, Peer Review) are activated.",
|
|
"coverage_score": 1.0,
|
|
"predicted_tier": "baseline",
|
|
"actual_best": "standard",
|
|
"match": false
|
|
},
|
|
{
|
|
"name": "Claude: University of Utah interview teaching example",
|
|
"subsample": "b",
|
|
"bucket": "high",
|
|
"frame_relationships": "The document discusses a university course, its curriculum, and related projects; these topics are interconnected as they all revolve around the teaching of computational design in a university setting.",
|
|
"coverage_score": 1.0,
|
|
"predicted_tier": "baseline",
|
|
"actual_best": "taxfree",
|
|
"match": false
|
|
},
|
|
{
|
|
"name": "ChatGPT: Starting Dropship Gun Business",
|
|
"subsample": "b",
|
|
"bucket": "high",
|
|
"frame_relationships": "The 'GunBusinessSetup' frame involves navigating multiple layers of compliance (FFL, zoning, and state/local laws) while building relationships with distributors and setting up an e-commerce platform.",
|
|
"coverage_score": 1.0,
|
|
"predicted_tier": "baseline",
|
|
"actual_best": "taxfree",
|
|
"match": false
|
|
},
|
|
{
|
|
"name": "ChatGPT: Analyze business plan",
|
|
"subsample": "b",
|
|
"bucket": "high",
|
|
"frame_relationships": "",
|
|
"coverage_score": 0.0,
|
|
"predicted_tier": "taxfree",
|
|
"actual_best": "taxfree",
|
|
"match": true
|
|
},
|
|
{
|
|
"name": "ChatGPT: Outdoor Layering Explained",
|
|
"subsample": "b",
|
|
"bucket": "mid",
|
|
"frame_relationships": "The frames interact by defining layers in an outdoor clothing system for managing sweat, insulation, and protection from weather conditions.",
|
|
"coverage_score": 1.0,
|
|
"predicted_tier": "baseline",
|
|
"actual_best": "taxfree",
|
|
"match": false
|
|
},
|
|
{
|
|
"name": "ChatGPT: Limits in Calculus.",
|
|
"subsample": "b",
|
|
"bucket": "mid",
|
|
"frame_relationships": "The 'Mathematics' and 'Calculus' frames are active. The 'Functions', 'Limits', and 'Calculus' frames are interconnected, as the document discusses the concept of limits in relation to functions within calculus.",
|
|
"coverage_score": 1.0,
|
|
"predicted_tier": "baseline",
|
|
"actual_best": "baseline",
|
|
"match": true
|
|
},
|
|
{
|
|
"name": "ChatGPT: Academic Program Director Role",
|
|
"subsample": "b",
|
|
"bucket": "mid",
|
|
"frame_relationships": "The 'academic program' frame underpins the entire document, with 'administration', 'faculty management', and 'student engagement' frames supporting it. The 'curriculum development' frame is a core function that intersects all other frames.",
|
|
"coverage_score": 1.0,
|
|
"predicted_tier": "baseline",
|
|
"actual_best": "taxfree",
|
|
"match": false
|
|
},
|
|
{
|
|
"name": "ChatGPT: Lonely Island Poop Skit",
|
|
"subsample": "b",
|
|
"bucket": "mid",
|
|
"frame_relationships": "The Music frame is used to deliver comedic content within the Humor frame.",
|
|
"coverage_score": 1.0,
|
|
"predicted_tier": "baseline",
|
|
"actual_best": "standard",
|
|
"match": false
|
|
},
|
|
{
|
|
"name": "ChatGPT: Parse Tidal playlist",
|
|
"subsample": "b",
|
|
"bucket": "mid",
|
|
"frame_relationships": "",
|
|
"coverage_score": 0.0,
|
|
"predicted_tier": "taxfree",
|
|
"actual_best": "taxfree",
|
|
"match": true
|
|
},
|
|
{
|
|
"name": "NO thesis proposal.pdf",
|
|
"subsample": "b",
|
|
"bucket": "document",
|
|
"frame_relationships": "The artist's career progression is intertwined with their identity and personal interests, shaping the subject matter and style of their product designs.",
|
|
"coverage_score": 1.0,
|
|
"predicted_tier": "baseline",
|
|
"actual_best": "baseline",
|
|
"match": true
|
|
},
|
|
{
|
|
"name": "PWM.pdf",
|
|
"subsample": "b",
|
|
"bucket": "document",
|
|
"frame_relationships": "The PWM and Analog to Digital Conversion frames are directly related as they both discuss the principles and applications of PWM in relation to analog-to-digital conversion, while the Arduino ADC frame provides specific context regarding the use of these concepts with an Arduino microcontroller.",
|
|
"coverage_score": 1.0,
|
|
"predicted_tier": "baseline",
|
|
"actual_best": "baseline",
|
|
"match": true
|
|
},
|
|
{
|
|
"name": "Will_It_Print.pdf",
|
|
"subsample": "b",
|
|
"bucket": "document",
|
|
"frame_relationships": "The document discusses the design and manufacturing of a part, focusing on its features, material removal, surface finish, support structures, overhangs, internal cavities, tolerances, and stress concentration.",
|
|
"coverage_score": 1.0,
|
|
"predicted_tier": "baseline",
|
|
"actual_best": "taxfree",
|
|
"match": false
|
|
},
|
|
{
|
|
"name": "Kim Kedem Ind Study F2025 Syllabus.docx",
|
|
"subsample": "b",
|
|
"bucket": "document",
|
|
"frame_relationships": "The Education frame serves as the context for the Project and Software Development frames, with the 3D Printing frame acting as a specific application of these skills.",
|
|
"coverage_score": 1.0,
|
|
"predicted_tier": "baseline",
|
|
"actual_best": "baseline",
|
|
"match": true
|
|
},
|
|
{
|
|
"name": "Aaron Nelson Graduate Transcript.pdf",
|
|
"subsample": "b",
|
|
"bucket": "document",
|
|
"frame_relationships": "",
|
|
"coverage_score": 0.0,
|
|
"predicted_tier": "taxfree",
|
|
"actual_best": "taxfree",
|
|
"match": true
|
|
}
|
|
]
|
|
} |