一举两得：字节精确的KV缓存嫁接将冻结的小模型转化为已验证知识的飞轮

摘要

我们报告了一种方法，可同时增强冻结小型语言模型的能力并大幅降低其成本，且无需调整任何权重。已验证的知识以字节精确的键值状态形式一次性存储，随后通过嫁接操作恢复至新的推理上下文中。该恢复过程为比特精确：在固定确定性配置下，嫁接后的逻辑输出与全新计算的结果逐字节一致（经SHA-256验证），零KL散度，且在五十个样本中实现100%的argmax一致。我们证明，在采用浮点旋转编码的模型上，自身位置嫁接是唯一数值精确的操作点，并在两种模型规模（12B、31B）及两个GPU目标上验证了字节精确性，其中一个目标通过预注册重放实现。在AIME 2025测试中，冻结的Gemma-4-12B模型在嫁接经过验证的解决方案库后，其性能从80.0%提升至93.3%，超越其自身77.5%的锚定值及其31B同类模型89.2%的已发布结果。在周期性任务中，基础模型在401,026个token预算内始终无法解决的八个问题，通过缓存的已验证解决方案仅用61个解码token完成回答，消耗token数减少6,574倍，能耗降低约8,700倍；其能力主张主要基于留存样本的迁移表现（31B模型上7/7全部成功）。相同的字节精确存储将可用上下文从32,768个token扩展至2,854,766个token，且无需额外加速器内存，并可在同构机器间实现字节一致迁移。我们从行为层面描述该系统；引擎为专有技术，所有报告数据均附有已提交的输入和输出哈希值，以便无需引擎即可重新验证评分结果。

English

We report a way to make a frozen small language model both more capable and dramatically cheaper at once, without changing any weights. Verified knowledge is deposited once as a byte-exact key-value (KV) state artifact and later restored, by graft, into a fresh inference context. The restore is bit-exact: under a pinned deterministic configuration, the grafted logits are byte-for-byte identical to a fresh computation (SHA-256 equality), with zero KL divergence and 100% argmax agreement over fifty samples. We show that own-position graft is the unique numerically exact operating point on a model with floating-point rotary encoding, and we verify byte-exactness on two model scales (12B, 31B) and two GPU targets, one through a pre-registered replay. On AIME 2025, a frozen Gemma-4-12B moves from 80.0% to 93.3% once a verified solution library is grafted, above its own 77.5% and its 31B sibling's 89.2% published anchors. On the recurring case, eight problems the base model never solves within a 401,026-token budget are answered from cached verified solutions in 61 total decode tokens, a factor of 6,574 fewer tokens and about 8,700x less energy; the capability claim proper rests on held-out transfer (7 of 7 at 31B). The same byte-exact store widens usable context from 32,768 to 2,854,766 tokens at zero extra accelerator memory, and moves byte-identical between machines of the same architecture. We describe the system at the behavior level; the engine is proprietary, and every reported number is backed by committed input and output hashes so the scoring can be re-checked without it.