name string | instruction string | description string | keywords list | schema_version string | config unknown | metadata unknown | files list | location string |
|---|---|---|---|---|---|---|---|---|
trial-digitalization/criteria-ALXN1840-WD-204 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-ALXN1840-WD-204",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "ALXN1840-WD-204",
"n_criteria": "31",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/ALXN1840-WD-204.json",
"tests/score.py",
"tests/test.sh"
] | criteria/ALXN1840-WD-204 | |
trial-digitalization/criteria-NCT00942578 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT00942578",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT00942578",
"n_criteria": "37",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT00942578.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT00942578 | |
trial-digitalization/criteria-NCT01247324 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT01247324",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT01247324",
"n_criteria": "47",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT01247324.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT01247324 | |
trial-digitalization/criteria-NCT01470612 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT01470612",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT01470612",
"n_criteria": "18",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT01470612.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT01470612 | |
trial-digitalization/criteria-NCT01492088 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT01492088",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT01492088",
"n_criteria": "35",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT01492088.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT01492088 | |
trial-digitalization/criteria-NCT01592370 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT01592370",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT01592370",
"n_criteria": "8",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT01592370.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT01592370 | |
trial-digitalization/criteria-NCT01624142 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT01624142",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT01624142",
"n_criteria": "28",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT01624142.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT01624142 | |
trial-digitalization/criteria-NCT01651403 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT01651403",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT01651403",
"n_criteria": "32",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT01651403.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT01651403 | |
trial-digitalization/criteria-NCT01711021 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT01711021",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT01711021",
"n_criteria": "32",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT01711021.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT01711021 | |
trial-digitalization/criteria-NCT01839487 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT01839487",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT01839487",
"n_criteria": "29",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT01839487.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT01839487 | |
trial-digitalization/criteria-NCT01881867 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT01881867",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT01881867",
"n_criteria": "37",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT01881867.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT01881867 | |
trial-digitalization/criteria-NCT01915498 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT01915498",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT01915498",
"n_criteria": "35",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT01915498.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT01915498 | |
trial-digitalization/criteria-NCT02047734 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02047734",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02047734",
"n_criteria": "48",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02047734.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02047734 | |
trial-digitalization/criteria-NCT02134028 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02134028",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02134028",
"n_criteria": "21",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02134028.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02134028 | |
trial-digitalization/criteria-NCT02164513 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02164513",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02164513",
"n_criteria": "38",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02164513.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02164513 | |
trial-digitalization/criteria-NCT02231749 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02231749",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02231749",
"n_criteria": "12",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02231749.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02231749 | |
trial-digitalization/criteria-NCT02264574 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02264574",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02264574",
"n_criteria": "33",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02264574.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02264574 | |
trial-digitalization/criteria-NCT02283294 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02283294",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02283294",
"n_criteria": "30",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02283294.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02283294 | |
trial-digitalization/criteria-NCT02336230 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02336230",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02336230",
"n_criteria": "27",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02336230.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02336230 | |
trial-digitalization/criteria-NCT02343120 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02343120",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02343120",
"n_criteria": "27",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02343120.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02343120 | |
trial-digitalization/criteria-NCT02354586 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02354586",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02354586",
"n_criteria": "29",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02354586.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02354586 | |
trial-digitalization/criteria-NCT02364999 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02364999",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02364999",
"n_criteria": "38",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02364999.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02364999 | |
trial-digitalization/criteria-NCT02401542 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02401542",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02401542",
"n_criteria": "41",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02401542.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02401542 | |
trial-digitalization/criteria-NCT02435212 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02435212",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02435212",
"n_criteria": "25",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02435212.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02435212 | |
trial-digitalization/criteria-NCT02435433 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02435433",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02435433",
"n_criteria": "41",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02435433.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02435433 | |
trial-digitalization/criteria-NCT02436668 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02436668",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02436668",
"n_criteria": "41",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02436668.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02436668 | |
trial-digitalization/criteria-NCT02485899 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02485899",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02485899",
"n_criteria": "10",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02485899.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02485899 | |
trial-digitalization/criteria-NCT02492711 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02492711",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02492711",
"n_criteria": "39",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02492711.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02492711 | |
trial-digitalization/criteria-NCT02520063 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02520063",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02520063",
"n_criteria": "39",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02520063.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02520063 | |
trial-digitalization/criteria-NCT02556788 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02556788",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02556788",
"n_criteria": "30",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02556788.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02556788 | |
trial-digitalization/criteria-NCT02561156 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02561156",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02561156",
"n_criteria": "35",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02561156.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02561156 | |
trial-digitalization/criteria-NCT02569541 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02569541",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02569541",
"n_criteria": "27",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02569541.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02569541 | |
trial-digitalization/criteria-NCT02586805 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02586805",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02586805",
"n_criteria": "14",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02586805.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02586805 | |
trial-digitalization/criteria-NCT02603120 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02603120",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02603120",
"n_criteria": "31",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02603120.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02603120 | |
trial-digitalization/criteria-NCT02607956 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02607956",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02607956",
"n_criteria": "30",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02607956.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02607956 | |
trial-digitalization/criteria-NCT02611817 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02611817",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02611817",
"n_criteria": "48",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02611817.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02611817 | |
trial-digitalization/criteria-NCT02630316 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02630316",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02630316",
"n_criteria": "20",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02630316.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02630316 | |
trial-digitalization/criteria-NCT02635776 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02635776",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02635776",
"n_criteria": "34",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02635776.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02635776 | |
trial-digitalization/criteria-NCT02642393 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02642393",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02642393",
"n_criteria": "28",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02642393.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02642393 | |
trial-digitalization/criteria-NCT02686658 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02686658",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02686658",
"n_criteria": "10",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02686658.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02686658 | |
trial-digitalization/criteria-NCT02709512 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02709512",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02709512",
"n_criteria": "33",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02709512.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02709512 | |
trial-digitalization/criteria-NCT02720094 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02720094",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02720094",
"n_criteria": "21",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02720094.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02720094 | |
trial-digitalization/criteria-NCT02732938 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02732938",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02732938",
"n_criteria": "32",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02732938.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02732938 | |
trial-digitalization/criteria-NCT02738138 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02738138",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02738138",
"n_criteria": "34",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02738138.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02738138 | |
trial-digitalization/criteria-NCT02781558 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02781558",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02781558",
"n_criteria": "32",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02781558.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02781558 | |
trial-digitalization/criteria-NCT02781571 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02781571",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02781571",
"n_criteria": "37",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02781571.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02781571 | |
trial-digitalization/criteria-NCT02821754 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02821754",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02821754",
"n_criteria": "28",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02821754.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02821754 | |
trial-digitalization/criteria-NCT02908672 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02908672",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02908672",
"n_criteria": "48",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02908672.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02908672 | |
trial-digitalization/criteria-NCT02915705 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02915705",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02915705",
"n_criteria": "26",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02915705.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02915705 | |
trial-digitalization/criteria-NCT02920476 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02920476",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02920476",
"n_criteria": "31",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02920476.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02920476 | |
trial-digitalization/criteria-NCT02926898 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT02926898",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT02926898",
"n_criteria": "31",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT02926898.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT02926898 | |
trial-digitalization/criteria-NCT03020160 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT03020160",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT03020160",
"n_criteria": "29",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT03020160.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT03020160 | |
trial-digitalization/criteria-NCT03032510 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT03032510",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT03032510",
"n_criteria": "26",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT03032510.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT03032510 | |
trial-digitalization/criteria-NCT03040999 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT03040999",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT03040999",
"n_criteria": "35",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT03040999.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT03040999 | |
trial-digitalization/criteria-NCT03056040 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT03056040",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT03056040",
"n_criteria": "29",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT03056040.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT03056040 | |
trial-digitalization/criteria-NCT03143153 | # Eligibility criteria extraction
`/app/protocol.pdf` is a clinical trial protocol. Extract **every eligibility
criterion** it states -- inclusion and exclusion -- and write them to
**`/app/output/criteria.json`**.
The document is the whole protocol, often more than a hundred pages, so finding
the eligibility section... | [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
] | 1.0 | {
"schema_version": "1.0",
"task": {
"name": "trial-digitalization/criteria-NCT03143153",
"keywords": [
"eligibility-criteria",
"clinical-trial",
"pdf",
"long-document",
"verbatim"
]
},
"metadata": {
"split": "test",
"difficulty": "medium",
"category": "docu... | {
"split": "test",
"difficulty": "medium",
"category": "document-understanding",
"study": "NCT03143153",
"n_criteria": "40",
"setting": "document"
} | [
"environment/Dockerfile",
"environment/inputs/protocol.pdf",
"instruction.md",
"solution/gold.json",
"solution/solve.sh",
"task.toml",
"tests/evaluator/criteria_match.py",
"tests/evaluator/evaluate.py",
"tests/gold/NCT03143153.json",
"tests/score.py",
"tests/test.sh"
] | criteria/NCT03143153 |
End of preview. Expand in Data Studio
TrialPD-Bench: Harbor tasks (test split)
Clinical trial protocol PDFs and expert-verified reference annotations, packaged as Harbor tasks so that a coding agent can be run and scored end to end. Only the test split is included; the protocols used to develop TrialScribe's skills and checklists are not in this dataset.
| Subset | Tasks | Input | Output | Headline metric |
|---|---|---|---|---|
indication/ |
40 | full protocol PDF | indications.json |
indication F1 |
criteria/ |
134 | full protocol PDF | criteria.json |
criterion F1 |
objectives/ |
89 | full protocol PDF | objectives.json |
objective / endpoint F1 |
soa_page/ |
86 | one SoA page (PDF) | soa_table.html |
Activity-Visit F1 |
soa_document/ |
37 | full protocol PDF | p<page>.html, one per SoA page |
Activity-Visit and Sub/sup F1 (missed pages score 0) |
timeline/ |
3 | full protocol PDF | timeline.json |
metric v2 (epoch accuracy, encounter / activity / timing F1) |
One task is one protocol (or, for soa_page, one page). manifest.csv lists all 389.
Layout
<subset>/<task_id>/
task.toml timeouts, resources; the verifier runs with no network
instruction.md the prompt, including the output contract
environment/
Dockerfile self-contained (python:3.11-slim + poppler + node)
inputs/ the PDF (and SPEC.md for the SoA subsets), baked in at /app
solution/ oracle: copies the gold into place
tests/ scorer, vendored evaluator, and the gold
manifest.csv
Run
harbor run -p criteria -a oracle -n 4 -o <jobs-dir> # should score 1.0
harbor run -p criteria -a claude-code -m <model> -n 4 -o <jobs-dir>
-o must be on local disk (Docker bind-mounts the verifier directory). The PDFs are
full protocols (median 122 pages), so each task image is large; use --limit while
iterating.
- Downloads last month
- 37