Small but Mighty: Test Suites as Improvement Signals for Self-Evolving Agents

Key points

• Verifying a rewrite by agent rollout is expensive and subjective, while executable checks such as the repository's own tests are too sparse to catch behavior-changing rewrites, rejecting only a quarter of them.1
• We instead take test feedback as the executable signal, and search for test suites that are both small and strong with a quality-diversity evolutionary algorithm, so that the signal is dense and informative.1
• TeSearch treats suite size as a behavior descriptor, so that quality is pursued at every size and a small, strong suite is reachable, and learns a transition kernel during the search that denoises the LLM as an actuator on that descriptor.1
• Across 129 real-world modules and three backbones, TeSearch delivers the highest-quality suites of any method compared, LLM-driven evolutionary search included, and the smallest of any LLM-based one.1

Sources (1)

• [1]Small but Mighty: Test Suites as Improvement Signals for Self-Evolving Agents
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 11, 10:36 PM

  A self-evolving agent rewrites its own harness and must decide which rewrites to keep, so its improvement signal is the check a rewrite has to pass.

  Verifying a rewrite by agent rollout is expensive and subjective, while executable checks such as the repository's own tests are too sparse to catch behavior-changing rewrites, rejecting only a quarter of them.

Extractive summary: sentences quoted from the sources.