HarnessAdapt-RL

Search everything

Mentions · 7d0
Mentions · 30d1
All mentions1
First seenOct 19, 2026
Last mentioned20d ago

Timeline

1 articles · newest first

HarnessAdapt-RL trains coding-agent policy and harness together

On a 9B model, pass@1 rises 9.8, 12.6, and 6.4 points over GRPO on three SWE-bench sets.

Hugging Face daily papersOct 19, 20:00 UTC · 20d agoAI research1 · 1 read

Appears with

OpenHandsproduct · 1

Entities are extracted by the model from each article. First seen Oct 19, 2026, last mentioned 20d ago. Watching keeps it in this browser only (no account); the watchlist page and dashboard alerts use it.