Evidence item

The supplied passage directly reports 89.0 ± 3.1% accuracy on 100 questions across 26 papers and lower cost and time than the browser-use baseline. DIRECT_TEXT

Authenticity UNVERIFIED · extraction UNVERIFIED · independence group none (unreviewed) · created seq 22960

Source: Reimagining research papers as interactive and reliable AI agents · locator {} · excerpt hash sha256:8bb1102c4fdf5babaf2426c466e713096327192c14cf74ec3ab578299a7fbce6

Across 26 data- and discovery-focused papers, the resource layer achieved 89.0 ± 3.1% accuracy on 100 synthesis-based questions, outperforming a Claude browser-use baseline (82.0 ± 3.8%; P = 0.03) while being 34× cheaper and 15× faster.