Microsoft 提出协同进化方法，将 Qwen3.5-35B-A3B 编码基准解决率从 57.8% 提升至 64.8%
Microsoft 这篇把 harness 和模型一起做协同进化，Qwen3.5-35B-A3B 三个编码基准解决率从 57.8% 拉到 64.8%，还能用 Opus 4.8 当提出者把经验迁移给小模型。
Microsoft 与合作者发布一项研究，通过让 harness（评测/执行框架）与模型协同进化来提升编码能力。在 Qwen3.5-35B-A3B 上，该方法把三个编码基准的平均解决率从 57.8% 提到 64.8%。流程是每轮由策略模型读取自身失败案例并提出 harness 修改，再用选定的 harness 进行 RL 训练，训练中持续困难的任务成为下一轮 harness 搜索集。协同优化的增益超过 harness 单独优化与策略单独优化增益之和，且各自优化后的部分与对方原始版本配对仍能保留提升。实验还显示用 Opus 4.8 担任 harness 提出者可带来额外增益，可将前沿模型的 harness 判断力迁移到更小的策略模型。
Great work from Microsoft and colleagues.
This one is on the hot new topic of co-evolving the harness and the model.
(bookmark it)
They raise Qwen3.5-35B-A3B's average resolve rate on three coding benchmarks from 57.8% to 64.8% by alternating the two.
In each round, the policy reads its own failures and proposes harness changes. Then it is trained with RL under the harness it picked. Tasks that stay hard during training become the search set for the next harness round.
The joint gain is larger than the harness-only gain and the policy-only gain added together. Each optimized part also keeps its improvement when paired with the other's original version.
Using Opus 4.8 as the harness proposer in the same loop adds more on top, which makes this a practical way to transfer a frontier model's harness judgment into a smaller policy.
https://t.co/FlYU8zQBCz
Chat with Paper: https://t.co/PrpXkk1qey
