腾讯混元发布 EvolveScaler 信息演化基准
🚀 EvolveScaler is here.
腾讯混元推出 EvolveScaler，一个把世界定义为可执行状态机、再渲染成自然语言的信息演化基准，包含 117 个原型、159 个问题算子和 5 个难度层级，单样本事件最多约 1,200 个。在最高难度层级上，14 个前沿模型的 median avg@5 降至 11.3；用它训练则在 8 个分布外基准上平均提升 5.25。
正文
当前显示原文，中文全文译文尚未就绪。
🚀 EvolveScaler is here.
Read a 40-day RPG log. Now answer one question: if you skip the mini-boss on Day 7, do you still beat the final boss?
The answer isn't in the log. You have to replay the world.
That's Information Evolution — records get retracted, corrected, backfilled. The world keeps changing after you read it.
So we build it backwards: define the world as an executable state machine, then render it into natural language. Code guarantees the logic. Language delivers the mess.
➡️ 117 prototypes. 159 question operators. 5 difficulty tiers. Up to ~1,200 events per sample.
➡️ 14 frontier models, hardest tier: median avg@5 falls to 11.3.
➡️ Train on it instead: +5.25 average across 8 out-of-distribution benchmarks.
Check out our paper and project page.