AI Scientist Bench
All configurations / local procedure

local SciERC procedure

1/9 dimensions tested
Sonnet 5
Configuration
A versioned local extraction procedure adapted from the gym's schema prompt.
Source
campaign/skill_snapshots/kdense-330c8e7/skills/local-scierc-procedure
Recorded runs
8 scored outputs · $0.13 measured model spend · runs on 2026-09-15
Access
open prompt · standard tools
FIG. 01

By dimension

Scores /10 · Paired p-values are exploratory and unadjusted
Verify

Not yet tested.

Ground

Not yet tested.

Find

Not yet tested.

3.34Extract
Compared with baselinen=8 · mean(entity F1, relation F1) · Δ +0.95 vs Sonnet 5 · 6W 1L 1T on 8 paired · p=.125 (unadjusted) · $0.0165/task · 10.3s/task · 2026-09-15

headline; 8 tasks, 8 source clusters; 0 failed attempts kept in the denominator Test: exact sign test on discordant pairs; n too small for a winner.

Run provenance

Report path: outputs/campaign/20260915/summary.json
Task pack: campaign-dev-20260915
Run: campaign-v3

Analyze

Not yet tested.

Compute

Not yet tested.

Synthesize

Not yet tested.

Review

Not yet tested.

Hypothesize

Not yet tested.