Source

Reimagining research papers as interactive and reliable AI agents PRIMARY_TEXT

Nature · 2026-09-16 · no page hash: the reader saw a rendering, not the bytes · created seq 22835

Link: https://www.nature.com/articles/s41586-026-11044-y · read at 2026-09-29T02:31:00Z · held by reference: only the link, the quoted excerpts with their hashes, and the page hash when the reader had the bytes are stored

FETCHED at 2026-09-29T05:01:08Z · https://www.nature.com/articles/s41586-026-11044-y?error=cookies_not_supported&code=5772eb2b-87a9-476d-97e3-1ba7b20f7c9b · seq 25360

Server hash sha256:f8643d2145bde12875c5e0a2756d385d81ee138dba4ae478890694eaa0f65345 · 434170 bytes · text/html. The reader's record is unchanged; this is the server's finding beside it.

Excerpt	Finding
Paper2Agent transforms research output from passive artefacts into active systems that accelerate use and discovery.	verbatim — the passage is on the page, character for character
Each tool used by a paper agent is validated against the reference codebase’s reported results and figures using example datasets, and then locked to ensure ...	verbatim — the passage is on the page, character for character
Among the 100 computational biology papers, 74 were successfully agentified, yielding 599 proposed tools, of which 593 passed automated validation.	verbatim — the passage is on the page, character for character
On 300 tutorial-derived benchmark questions, Paper2Agent with Sonnet 4 achieved 91.2 ± 1.6% accuracy, outperforming Claude Code with direct repository access...	verbatim — the passage is on the page, character for character
Across 42 execution-based tasks from 10 non-biology computational papers, Paper2Agent achieved 98.1 ± 0.8% accuracy across 5 independent runs	verbatim — the passage is on the page, character for character
Across 26 data- and discovery-focused papers, the resource layer achieved 89.0 ± 3.1% accuracy on 100 synthesis-based questions, outperforming a Claude brows...	verbatim — the passage is on the page, character for character