ThinkJev：一次思考覆盖所有预算的校准型决策
原标题：ThinkJev: One Thought, Every Budget for Calibrated Typed Decisions
AI 摘要
研究者提出 ThinkJev，一种无需训练、基于单一混合思考模型（如 Qwen3 系列）的级联决策方法：System 1 直接读取选项字母概率作答，仅在统计检验保证错误率低于目标阈值时单独回答；其余决策升级到 System 2，在已缓存的同一提示上继续思考，并可在任意思考预算下读取同一答案槽的概率。在 8359 个测试决策上，Qwen3.5-27B 以 5% 目标错误率让 System 1 单独回答 24.5% 的决策，实际错误率 3.9%；其余决策在答案置信时提前停止思考，达到 81.6% 准确率，仅用 72% 的思考 token。
正文节选
ThinkJev: One Thought, Every Budget for Calibrated Typed Decisions Abstract Many uses of language models end in a typed decision: a choice among a closed set of options, whose probabilities code can act on. Reading those probabilities from one forward pass is fast but often wrong on decisions that take several steps, such as arithmetic, counting, and date comparison. Letting the model think first fixes most of these errors, but costs seconds per decision instead of milliseconds. To pay for think