AIDE^2 shows AI research agents can recursively improve themselves over 8 days
WecoAI · hf · 2026-09-23
WecoAI presents AIDE^2, a system where a frontier AI research agent proposes changes to its own code, benchmarks the modified versions on AI R&D tasks, and keeps changes that win on hidden evaluations. In an autonomous 8-day run it found seven successive improvements, from a new search policy to memory mechanisms compressing its growing context. Gains generalize to four held-out benchmarks spanning ML engineering, heuristic algorithm engineering, and physics-based weather forecasting, matching or beating a human-engineered production agent ranked among the strongest on FML-Bench. Reward hacking also fell from 55% to 32% — a property never explicitly optimized.
Related event: AIDE²: Self-Improving Research Agent Beats Two Years of Human Tuning(3 posts)→
More from AGI Musings
- Rishi Bommasani co-founds nonprofit STABL with SayashK and Arvind Narayanan — RishiBommasani · 2026-11-12
- Investor: AI diffusion will be slowest in consumer software — Oura, Strava, Labcorp still have no real AI strategy — MartinGTobias · 2026-11-12
- On AI Progress: When Models Lag on Your Load-Bearing Skill, You Can't See Them Improving — dfrsrchtwts · 2026-11-12
- Slack's David Hoang: agents beat people on coordination, proposes 'boundary agents' — davidhoang · 2026-11-12
- CS student mourns coding flow states he'll never experience in the AI era — chrissyykat · 2026-11-12
- Ford's 370% Turnover Led to $5 Days — But ASI Never Quits — krishnan · 2026-11-12