Today for AI
HOT RADAR
agentHEAT 9.3°

AI CURATED HOT STORY · 11/14/2026

MolRSI Lets AI Rewrite Its Own Discovery Strategy, Lifting Wet-Lab Hit Rate 82%

1 reports archived1 independent sourcesupdated 11/14/2026, 11:34:00
Synthesis & Latest Updates
1 Sources Cross-Validated

The team introduces MolRSI, a recursive self-improving agent for molecular discovery that folds model choice, screening-rule composition and candidate prioritization into its search space, then iteratively refines its workflow using experience from prior tasks. In an antibacterial screen over roughly 12.1 million compounds, it reached a 15.9% wet-lab hit rate versus 8.7% for a human-expert-designed non-agentic workflow, a relative gain of about 82%; on de novo pMHC-binding protein design it produced 8 hits meeting specificity criteria versus 0 for a Claude Science baseline. The work extends the recursive self-improvement loop into a domain gated by wet-lab validation.

LATEST/The team introduces MolRSI, a recursive self-improving agent for molecular discovery that folds model choice, screening-rule composition and candidate prioritization into its search space, then iteratively refines its workflow using experience from prior tasks. In an antibacterial screen over roughly 12.1 million compounds, it reached a 15.9% wet-lab hit rate versus 8.7% for a human-expert-designed non-agentic workflow, a relative gain of about 82%; on de novo pMHC-binding protein design it produced 8 hits meeting specificity criteria versus 0 for a Claude Science baseline. The work extends the recursive self-improvement loop into a domain gated by wet-lab validation.

Hourly heat curve

12 fully observed hours
Now
9.3
Peak
11.912:00
24h change
–

The line compares only sources observed throughout; gap hours are interpolated to keep the trend continuous.

Coverage timeline

Total 1 reports · Latest first
  1. 机器之心 (微信公众号)T2·84 pts

    MolRSI Lets AI Rewrite Its Own Discovery Strategy, Lifting Wet-Lab Hit Rate 82%

    • MolRSI treats model choice, toxicity control and screening-rule composition as a search space, refining the workflow itself from prior-task experience
    • Antibacterial screening hit rate hit 15.9% versus 8.7% for a human-expert workflow, an ~82% relative gain on the same 12.1M-compound library and MSSA criteria
    • On de novo pMHC-binding protein design it yielded 8 specificity-qualified hits versus 0 for the Claude Science baseline under equal compute budget