Dataset research report
Natural Questions research report
A reproducible data report with schema notes, generated chart evidence, suggested follow-up questions, and export-ready Helix queries.
Executive Summary
Dataset Card for Natural Questions Dataset Summary The NQ corpus contains questions from real users, and it requires QA systems to read and comprehend an entire Wikipedia article that may or may not contain the answer to the question. The inclusion of real user questions, and the requirement that solutions should read an entire page to find the answer, cause NQ to be a more realistic and challenging task than prior QA datasets. Supported Tasks and Leaderboards… See the full description on the dataset page: https://huggingface.co/datasets/google-research-datasets/natural_questions.
Finding 1The dataset has 500 rows available in the catalog.
Finding 2The catalog exposes 5 documented or inferred columns.
Finding 3Helix has 4 ready query prompts for this dataset.
Finding 4This report still exposes schema, preview rows, and query prompts even when charts cannot be precomputed.
Follow-Up Queries
Preview Rows
| # | idtext | documenttext | questiontext | long_answer_candidatestext | annotationstext |
|---|---|---|---|---|---|
| 1 | 4549465242785278785 | {'html': '<!DOCTYPE html>\n<HTML class="client-js ve-not-available" lang="en" dir="ltr"><HEAD>\n\n<TITLE>The Walking Dead (season 8) - Wiki… | {'text': 'when is the last episode of season 8 of the walking dead', 'tokens': array(['when', 'is', 'the', 'last', 'episode', 'of', 'season… | {'end_byte': array([57620, 53883, 54388, 56148, 56129, 55557, 56249, 56364, 56522, 56674, 56962, 57130, 57392, 57603, 58549, 59539, … | {'id': array(['6782080525527814293'], dtype=object), 'long_answer': array([{'candidate_index': 92, 'end_byte': 96948, 'end_token': 3538, 's… |
| 2 | -2543388002166163252 | {'html': '<!DOCTYPE html>\n<HTML class="client-js ve-not-available" lang="en" dir="ltr"><HEAD>\n\n<TITLE>Persephone - Wikipedia</TITLE>\n\n… | {'text': 'in greek mythology who was the goddess of spring growth', 'tokens': array(['in', 'greek', 'mythology', 'who', 'was', 'the', 'godd… | {'end_byte': array([62898, 58933, 59089, 59894, 60156, 60254, 60446, 60618, 60763, 62753, 62882, 63581, 63565, 81194, 64068, 64298, … | {'id': array(['7719528322202775345'], dtype=object), 'long_answer': array([{'candidate_index': 58, 'end_byte': 84070, 'end_token': 965, 'st… |
| 3 | 5985355041383167183 | {'html': '<!DOCTYPE html>\n<HTML class="client-js ve-not-available" lang="en" dir="ltr"><HEAD>\n\n<TITLE>Colony (biology) - Wikipedia</TITL… | {'text': 'benefits of colonial life for single celled organisms', 'tokens': array(['benefits', 'of', 'colonial', 'life', 'for', 'single', '… | {'end_byte': array([54151, 54966, 55845, 59144, 60541, 59465, 59860, 60535, 60529, 60523, 61333, 62062, 63676, 64953, 65246, 66028, … | {'id': array(['13676402902866580638'], dtype=object), 'long_answer': array([{'candidate_index': -1, 'end_byte': -1, 'end_token': -1, 'start… |
| 4 | -2975172535563055798 | {'html': '<!DOCTYPE html>\n<HTML class="client-js ve-not-available" lang="en" dir="ltr"><HEAD>\n\n<TITLE>The Man in the High Castle (TV ser… | {'text': 'how many season of the man in the high castle', 'tokens': array(['how', 'many', 'season', 'of', 'the', 'man', 'in', 'the', 'high'… | {'end_byte': array([49862, 43135, 44219, 44200, 44344, 44592, 45535, 45516, 45073, 45173, 45425, 45787, 45986, 45967, 46059, 46157, … | {'id': array(['7446307064203576492'], dtype=object), 'long_answer': array([{'candidate_index': 0, 'end_byte': 49862, 'end_token': 473, 'sta… |
| 5 | -1052334833502528495 | {'html': '<!DOCTYPE html>\n<HTML class="client-js ve-not-available" lang="en" dir="ltr"><HEAD>\n\n<TITLE>List of heads of state of Nigeria … | {'text': 'who was the first ministry head of state in nigeria', 'tokens': array(['who', 'was', 'the', 'first', 'ministry', 'head', 'of', 's… | {'end_byte': array([48170, 42667, 43557, 46355, 46343, 43856, 43850, 44088, 44082, 44283, 44277, 44828, 44822, 44816, 45382, 45376, … | {'id': array(['3569531263672159632'], dtype=object), 'long_answer': array([{'candidate_index': -1, 'end_byte': -1, 'end_token': -1, 'start_… |
| 6 | -6252343352866892945 | {'html': '<!DOCTYPE html>\n<HTML class="client-js ve-not-available" lang="en" dir="ltr"><HEAD>\n\n<TITLE>List of awards and nominations rec… | {'text': 'how many nominations does game of thrones have', 'tokens': array(['how', 'many', 'nominations', 'does', 'game', 'of', 'thrones', … | {'end_byte': array([72681, 71528, 71516, 56981, 57278, 57727, 57478, 57472, 58144, 57895, 57889, 58548, 58299, 58952, 58703, 59399, … | {'id': array(['14840975513360924403'], dtype=object), 'long_answer': array([{'candidate_index': -1, 'end_byte': -1, 'end_token': -1, 'start… |
Data Dictionary
- id text
- document mixed
- question mixed
- long_answer_candidates datetime
- annotations mixed
Method And Limits
- Load the catalog entry and preview rows from the processed dataset file.
- Infer numeric, categorical, time, and location fields from real columns.
- Generate a small set of defensive Plotly chart specifications from that profile.
- Expose each chart idea as a query link so the report can be rerun or exported in Helix.
This report is intentionally reproducible. It uses the local catalog metadata and generated chart specifications rather than claiming external conclusions beyond the dataset.