Peng Xia, Rujun Han, Zifeng Wang, Yanfei Chen, Yufan Zhang, Yoonho Lee, Chengsong Huang, Han Yu, Zhongying CuiZhu, Yifei Ming, Huaxiu Yao, Burak Gokturk, Tomas Pfister, Chen-Yu Lee — Google Cloud AI Research
Charts below re-plot numbers reported in the paper (Figures 1 & 3, Tables 1–3). Policy frozen at Claude Opus 4.8; the harness is evolved on one suite per domain and then evaluated unchanged on held-out and out-of-distribution (OOD) benchmarks.
Agentic workspace instance. Points far to the upper-left are the goal: small evolve-set gain, large transfer. Points left of the dashed line at 0 fall below the unevolved harness H₀.
Reading it: Meta-Harness has the biggest evolve-set gain (+3.6) but keeps only +0.9 OOD. AHE and TTHE are negative OOD (−0.5 and −1.7) — their evolution actively hurt tasks they were never scored on. RRSI has the smallest evolve gain (+1.1) and the largest transfer (+3.9), the exact trade the regularizers are designed to make.
Every held-out split improves and none regresses. Frontier-Eng is measured in medal points: +4.3 is a 24.3% relative gain. The engineering-design domain matters because its tasks are graded by frozen deterministic simulators, not a judge model — so the transfer can't be explained by a harness learning to write the way a judge rewards.
Regularization is cheaper, not more expensive: RRSI spends 2.42M tokens/trial against 3.80M for unregularized evolution (≈36% less), and only 1.6× the unevolved baseline. Unregularized selection spends most of its accepted edits on noise and on context rather than on mechanism — it buys score with compute, and the compute doesn't transfer.
| Method | In-distribution | Out-of-distribution | OOD avg. | |||
|---|---|---|---|---|---|---|
| Harvey LAB (evolve) | Harvey LAB (ID held-out) | JobBench | GDPval | APEX-Agents | ||
OOD avg. is computed as the mean of JobBench, GDPval and APEX-Agents (the paper reports only the H₀ and RRSI averages explicitly, 39.7 and 43.6; the baseline averages shown for the others are recomputed from the table's own columns). Same H₀, same evolve split, same candidate budget for every row.
| Variant | Harvey LAB (evolve) | Harvey LAB (ID held-out) | OOD avg. | Tokens/trial (M) ↓ |
|---|
The strongest single signal in the paper: evolve-set score and transfer move in opposite directions. Dropping both regularizer groups yields the highest evolve score of any arm (92.8) and an OOD average within a point of the unevolved harness. Dropping only the acceptance constraints raises the evolve score by 1.0 pt, costs 2.6 pts OOD, and inflates token cost by half.
| Policy | Benchmark | H₀ | RRSI | Δ |
|---|
Evolution is run independently per policy on Terminal-Bench 2.1 only; the resulting harness is then evaluated unchanged on SWE-bench Verified. Gemini 3.5 Flash gains more on the evolve split (+14.1 — the paper's headline number) while the stronger Claude Opus 4.8 sits closer to both ceilings and gains less.