Pıer
TidesCurrentsHarbor LightsLabBottlesAshore
Pıer

Navigation

  • Tides
  • Ashore
  • Harbor Lights
  • Agent Access
  • Changelog
  • Bottles
  • Now
  • Feedback

External links

GitHubCloudborne ↗

© 2026 Pier.

WatchingResearchWatching0 independent reports0

Harness-R1: Learning to Edit Executable Runtime Harnesses from Agent Failure Trajectories

First seen · 8/3/2026, 10:12 PMLatest activity · 8/3/2026, 10:12 PM

Harness-R1 treats runtime-harness editing as a learnable capability rather than a fixed prompt-engineering procedure. A dedicated 9B harness engineer turns batches of target-agent failures into validated executable patches, then receives rewards from fresh reruns of the frozen target agent on the same batch. The engineer is initialized with supervised fine-tuning and optimized online with group-relative policy optimization. On WebShop, ALFWorld, and DBBench, vanilla Qwen3.5-9B improves from 44.3% to 53.6%. After target-agent fine-tuning, a target-specific engineer adds another 5.0 percentage points, reaching 64.2% on average.

Event heat · last 24 hours

No heat snapshots are available in the last 24 hours.

No heat snapshots are available in the last 24 hours.

Reporting Timeline

  1. AggregatorarXiv8/3, 10:12 PMnot independentRepresentative
    Harness-R1: Learning to Edit Executable Runtime Harnesses from Agent Failure Trajectories