Export
Machine-readable JSONL, JSON, or CSV for AI training and reuse, a human-readable XLSX reproducing the column groups, and a printable PDF report — or export the paper metadata alone, one row per paper, or every attached PDF as a .zip. All formats respect the filters below — alloy type, search, field state, and which metadata columns and data points to include.
Alloy-type filter
All 0 papers · 0 experiments included.
Paper metadata to include
These bibliographic columns appear in every format below, and are the columns used by the metadata-only export. Turn any off to leave it out.
Columns & data points to include
These toggles control which experiment data points appear in every format below. Paper metadata columns are chosen separately, above.
Full dataset
One row/record per experiment.
JSONL
One record per experiment, typed, with explicit states and a schema header line.
CSV
Flat one-row-per-experiment table with paper columns — opens anywhere, re-imports here.
JSON
A single { schema, records } document — same typed records, pretty-printed.
XLSX
Column-group layout, one row per experiment, values human-readable.
PDF report
Formatted report grouped by alloy type — opens the print dialog (save as PDF).
Metadata only
One row/record per paper — just the selected bibliographic columns, no experiment data. 0 papers · 16 columns.
Metadata CSV
One row per paper with the chosen metadata columns — opens anywhere, re-imports here.
Metadata XLSX
A single sheet of paper metadata, one row per paper.
Metadata JSON
A { columns, papers } document of the selected metadata, one object per paper.
PDF files
The uploaded PDFs themselves — bundled into one .zip, named by citation key. 0 of 0 filtered papers have a PDF attached.
Fetches each attached PDF and packages them into a single archive, downloaded in your browser.