developer tools
The Supervisor Could Interrupt the Run
PILOT separates a long-running worker from a supervisor that can steer or abort it and retain lessons as skills.
Summary
PILOT separates a long-running worker from a supervisor that can steer or abort it and retain lessons as skills.
The harness couples live steering with live self-evolution, distilling procedures and failure modes while a task is still active. Across two frozen model backbones and three benchmarks, the authors report first place in five of six configurations, gains up to 9.8 points on Terminal-Bench 2.0 and lower output-token use. In self-improvement settings they report 12.4- and 14.6-point gains. These are benchmarked harness effects, not evidence that intervention always improves open-ended work.
Why it matters
PILOT separates a long-running worker from a supervisor that can steer or abort it and retain lessons as skills.
Limits and context
- These are benchmarked harness effects, not evidence that intervention always improves open-ended work.
Key claims
PILOT separates a long-running worker from a supervisor that can steer or abort it and retain lessons as skills.
Qualification: These are benchmarked harness effects, not evidence that intervention always improves open-ended work.
Evidence: source-2026-08-30-013
Sources
- arXiv preprint 2608.26530arXiv · primary research
Corrections
No corrections have been recorded for this story.