SanSi 用循环层重算让小模型决策更准：72.0% 对比 58.4%
遇到只能二选一的分类调用时，这篇给了个省钱思路：1.4B 小模型循环跑 8 次层，比单次的 4.2B 模型还准。
论文提出 SanSi，将预训练的 Ouro-1.4B 模型的同一组层循环运行最多 8 次，每次循环后读取选项概率，且每次循环都训练到能给出答案，因此可在 1 到 8 次循环间灵活停止。在 10,027 条测试决策上，SanSi 得分 72.0%，同结构不循环模型为 58.4%，计算量为 7.7 倍。在比训练时更长的 liar 链上，SanSi 达到 72.6%，而 4.2B 单次推理模型只有 50.0%，等于随机水平。
This paper finds that small decision models get far more accurate by rerunning the same layers several times before giving an answer.
Many LLM calls only need a choice, like spam or not spam. Typed decision models answer these in a single pass, which fails when a decision needs several dependent steps.
SanSi loops Ouro-1.4B, a model pretrained to reuse its layers, and reads option probabilities after each of 8 loops. Every loop is trained to answer, so the same model can stop anywhere from 1 to 8 loops.
On 10,027 test decisions, SanSi scored 72.0% versus 58.4% for a same-shape model without loops, at 7.7x the compute. On liar chains longer than any seen in training, it reached 72.6% while a 4.2B single-pass model scored 50.0%, the chance level.
If memory limits your decision-style calls, try a looped small model before moving to a model 3x larger.
