Question-and-answer datasets for research and AI

8 extractive question-and-answer pairs built from 1725952345 neurology dementia, published by icmr.gov.in. The source is a 1-page publication. Source-evidenced topics include dementia, cognitive decline, depression, delirium, functional decline, probable ad, mixed dementia, lab investigation. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 3 list, 2 definition, 2 factual, 1 summary. 6 of the 8 pairs (75.0%) ask explanatory questions; the remaining 2 are factual or table-based questions. 100.00% of rows pass the corpus quality gate.

Rows

8

Format

CSV

Licence

Not specified

Updated

4 Oct 2026

11 extractive question-and-answer pairs built from Uterine Fibroids and Polyps, published by icmr.gov.in. The source is a 1-page publication. Source-evidenced topics include standard treatment workflow, uterine fibroids, polyps, icd-11-2e86.0, india, icmr, secondary care hospitals, tertiary care hospitals. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 7 list, 3 factual, 1 relationship. 8 of the 11 pairs (72.7%) ask explanatory questions; the remaining 3 are factual or table-based questions. 100.00% of rows pass the corpus quality gate.

Rows

11

Format

CSV

Licence

Not specified

Updated

4 Oct 2026

18 extractive question-and-answer pairs built from Fever In Children, published by icmr.gov.in. The source is a 1-page publication. Source-evidenced topics include fever, rash, joint symptoms, soft tissue swelling, redness, axillary temperature, general examination, macules. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 7 definition, 7 list, 4 relationship. 18 of the 18 pairs (100.0%) ask explanatory questions; the remaining 0 are factual or table-based questions. 100.00% of rows pass the corpus quality gate.

Rows

18

Format

CSV

Licence

Not specified

Updated

4 Oct 2026

17 extractive question-and-answer pairs built from Severe Acute Malnutrition, published by icmr.gov.in. The source is a 1-page publication. Source-evidenced topics include severe acute malnutrition, standard treatment workflow, icd-10-e43, loose motions, jaundice, seizures, pneumonia, diarrhea. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 9 factual, 6 list, 1 definition, 1 summary. 8 of the 17 pairs (47.1%) ask explanatory questions; the remaining 9 are factual or table-based questions. 100.00% of rows pass the corpus quality gate.

Rows

17

Format

CSV

Licence

Not specified

Updated

4 Oct 2026

28 extractive question-and-answer pairs built from Standard Treatment Workflow on Paediatric Intrathoracic Tuberculosis, published by icmr.gov.in. The source is a 1-page publication. Source-evidenced topics include weight loss, persistent fever, unremitting cough, chest x-ray, paratracheal lymph nodes, fibrocavitory disease, miliary pattern, pleural effusion. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 11 factual, 9 definition, 5 list, 3 relationship. 17 of the 28 pairs (60.7%) ask explanatory questions; the remaining 11 are factual or table-based questions. 100.00% of rows pass the corpus quality gate.

Rows

28

Format

CSV

Licence

Not specified

Updated

4 Oct 2026

11 extractive question-and-answer pairs built from Standard Treatment Workflow on Paediatric Lymph Node Tuberculosis, published by icmr.gov.in. The source is a 1-page publication. Source-evidenced topics include paediatric lymph node tuberculosis, standard treatment workflow (stw), icd-10-a18.2, chronic sinus, cold abscess, chest x-ray, fnac, naat. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 8 definition, 2 list, 1 relationship. 11 of the 11 pairs (100.0%) ask explanatory questions; the remaining 0 are factual or table-based questions. 100.00% of rows pass the corpus quality gate.

Rows

11

Format

CSV

Licence

Not specified

Updated

4 Oct 2026

15 extractive question-and-answer pairs built from Standard Treatment Workflow on Paediatric Osteoarticular Tuberculosis, published by icmr.gov.in. The source is a 1-page publication. Source-evidenced topics include standard treatment workflow (stw), paediatric osteoarticular tuberculosis, national tb elimination programme, dactylitis, arthritis, x-ray, fnac, naat. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 8 definition, 5 list, 2 factual. 13 of the 15 pairs (86.7%) ask explanatory questions; the remaining 2 are factual or table-based questions. 100.00% of rows pass the corpus quality gate.

Rows

15

Format

CSV

Licence

Not specified

Updated

4 Oct 2026

110 extractive question-and-answer pairs built from Standard Treatment Workflows of India on Paediatric and Extrapulmonary Tuberculosis, published by icmr.gov.in. The source is a 16-page publication. Source-evidenced topics include paediatric tuberculosis, extrapulmonary tuberculosis, national tb elimination programme, standard treatment workflows, central tb division, paediatric abdominal tuberculosis, paediatric intrathoracic tuberculosis, paediatric lymph node tuberculosis. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 38 factual, 33 list, 26 definition, 8 relationship, 4 summary, 1 comparison. 72 of the 110 pairs (65.5%) ask explanatory questions; the remaining 38 are factual or table-based questions. 100.00% of rows pass the corpus quality gate.

Rows

110

Format

CSV

Licence

Not specified

Updated

4 Oct 2026

24 extractive question-and-answer pairs built from ICMR Standard Treatment Workflow (STW) CUTANEOUS ADVERSE DRUG REACTIONS - PART A, published by icmr.gov.in. The source is a 1-page publication. Source-evidenced topics include cutaneous adverse drug reactions (cadr), urticaria, angioedema, dress, stevens johnson syndrome, toxic epidermal necrolysis, fixed drug eruption (fde), maculopapular/exanthematous reactions. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 10 factual, 10 list, 3 summary, 1 definition. 14 of the 24 pairs (58.3%) ask explanatory questions; the remaining 10 are factual or table-based questions. 100.00% of rows pass the corpus quality gate.

Rows

24

Format

CSV

Licence

Not specified

Updated

4 Oct 2026

4 extractive question-and-answer pairs built from ICMR Standard Treatment Workflow (STW) DERMATOPHYTOSES, published by icmr.gov.in. The source is a 1-page publication. Source-evidenced topics include dermatophytoses, standard treatment workflow, icd-10-b35.9, clotrimazole, miconazole, terbinafine, ketoconazole, oral antifungals. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 2 definition, 1 list, 1 relationship. 4 of the 4 pairs (100.0%) ask explanatory questions; the remaining 0 are factual or table-based questions. 100.00% of rows pass the corpus quality gate.

Rows

4

Format

CSV

Licence

Not specified

Updated

4 Oct 2026

21 extractive question-and-answer pairs built from ICMR Standard Treatment Workflow (STW) LIVER FAILURE, published by icmr.gov.in. The source is a 1-page publication. Source-evidenced topics include liver failure, standard treatment workflow, acute liver failure, aclf, diagnostic criteria, management, primary treatment, stabilization. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 10 factual, 8 definition, 2 list, 1 summary. 11 of the 21 pairs (52.4%) ask explanatory questions; the remaining 10 are factual or table-based questions. 100.00% of rows pass the corpus quality gate.

Rows

21

Format

CSV

Licence

Not specified

Updated

4 Oct 2026

23 extractive question-and-answer pairs built from Image guided management of Obstructive Jaundice, published by icmr.gov.in. The source is a 1-page publication. Source-evidenced topics include cholangitis, obstructive jaundice, biliary drainage, percutaneous transhepatic biliary drainage (ptbd), ercp, usg abdomen, mrcp, cect abdomen. Every answer is a verbatim span of text the source prints, and each row carries the passage it sits in, its offset in that passage, the source quote, the page and the location in the document, so any row can be checked against the original. Question types in the export: 14 list, 5 factual, 2 relationship, 2 summary. 18 of the 23 pairs (78.3%) ask explanatory questions; the remaining 5 are factual or table-based questions. 100.00% of rows pass the corpus quality gate.

Rows

23

Format

CSV

Licence

Not specified

Updated

4 Oct 2026