Preprint
Sep 2026
A Theory of Reliable Self-Evolution for Agent Harnesses
In harness self-evolution, agents modify their own prompts, code, tools, and orchestration while keeping the underlying language model fixed. Recent work has shown that agents can improve themselves in response to task failures and achieve substantial performance gains. However, gains on failed tasks do not automatical...
Qi Cai, Yong-Gang Zhang, Jun Nie et al.
· 2 citations
· ⚡2