Where Does Harness-Optimization Value Live? Localized Gains and the Budget-Splitting Trap in Self-Evolving LLM Agents

The paper introduces HARNESSEVO, a method that decomposes a large language model’s harness into four independently evolvable components—role, task‑strategy, tool/format‑rules, and reflection/control. Experiments on ALFWorld show that overall success rates are similar to flat‑string evolution, but the reflection/control component alone accounts for most of the performance gains. The study also finds that evenly distributing optimization budget across all slots can be detrimental; concentrating resources on the high‑credit control slot recovers lost performance, while on WebShop all slots remain ineffective, suggesting task‑specific differences in harness value.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computation and Language.

• llms
• agents