← Back to all news
arXiv AI By Jingzhe Liu, Jiliang Tang, Tong Zhao, Neil Shah, Clark Mingxuan Ju

Forging More Tasks into Better Harnesses: Agentic Harness Evolution via Successive Scale Racing

Read the original on arXiv AI →

The paper introduces Successive Scale Racing (SSR), a framework that improves agentic harness evolution by adaptively selecting optimal evidence combinations during each evolution round. SSR constructs diverse evidence sets based on task quality and coverage, then efficiently races candidate harnesses under a limited compute budget. Across three benchmarks, SSR outperforms baselines, steadily raising pass rates as the task pool expands from 40 to 200 while cutting validation executions by 60%.

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 AI.

Related stories

arXiv AI
Oct 1

Self-Evolving Harness on Multiple Tasks with the Agent as Its Own Optimizer

The paper introduces a self‑evolving harness framework where a frozen language‑model agent first solves tasks and then edits its own harness based on run records. Using a 49‑line seed harness, the evolved harness improves average scores on in‑distribution benchmarks by 4.48 points and on out‑of‑distribution benchmarks by 12.64 points, surpassing Codex on the former and matching it on the latter. Continued evolution on a specific out‑of‑distribution benchmark further raises performance, and the study analyzes emergent mechanisms such as output truncation and history compaction.

By Qiankai Xu
agentsbenchmarks
More like this →
arXiv AI
Aug 28

Verify Smarter, Evolve Further: Efficient Harness Evolution through Behavior-Aware Verification

The paper introduces HarnessLens, a budget‑aware framework that evolves language‑model agent harnesses by jointly exploring task spaces and user‑configurable components. It derives candidate modifications from execution trajectories and selectively verifies them on behavior‑relevant tasks using an attributable‑evidence gate, thereby avoiding wasteful rollouts on unrelated behaviors. Experiments on three harnesses and four benchmarks show that HarnessLens improves held‑out performance by 7.6‑13.6% while using less evaluation budget than existing methods.

By Jinghan Xu, Yikai Zhang, Aili Chen, Weiyuan Li, Jiaqing Liang, Deqing Yang
agentsbenchmarks
More like this →