SearchJev：面向搜索智能体的快速且校准的System-1模型

中文摘要 AI 辅助

搜索智能体反复就相关性、证据充分性和搜索动作做出短暂决策。使用生成式语言模型进行这些决策会引入延迟和不可靠的置信度。我们提出SearchJev，一个快速且校准的System-1模型，它将搜索决策与System-2推理和生成分离。给定搜索状态和决策模式，SearchJev直接对合法选项进行评分，无需自回归输出生成。我们提出软标签学习用于校准决策（SLCD），以从不确定的监督中学习决策概率并校准其置信度。在双系统搜索智能体中，SearchJev处理短暂决策，并将不确定的判断委托给System-2，后者保留规划、查询生成和答案组合。我们还引入了SearchDecision-Bench，一个统一六种搜索决策类型用于训练和评估的基准。在SearchDecision-Bench上，SEARCHJEV在决策质量上优于同尺寸的Qwen3.5自回归模型，决策速度提高5.2-5.3倍，平均期望校准误差降低41-74%。在BrowseComp-Plus上，双系统智能体在主动搜索时间上实现3.7-4.7倍的加速，同时将答案准确率从45%提升至最高54%。

英文摘要

Search agents repeatedly make short decisions about relevance, evidence sufficiency, and search actions. Using generative language models for these decisions introduces latency and unreliable confidence. We present SearchJev, a fast and calibrated System-1 model that separates search decisions from System-2 reasoning and generation. Given a search state and a decision schema, SearchJev directly scores legal options without autoregressive output generation. We propose Soft-Label Learning for Calibrated Decisions (SLCD) to learn decision probabilities from uncertain supervision and calibrate their confidence. In a dual-system search agent, SearchJev handles short decisions and delegates uncertain judgments to System 2, which retains planning, query generation, and answer composition. We also introduce SearchDecision-Bench, a benchmark unifying six types of search decisions for training and evaluation. On SearchDecision-Bench, SEARCHJEV improves decision quality over same-size Qwen3.5 autoregressive models, achieves 5.2-5.3 times faster decisions, and reduces average expected calibration error by 41-74%. On BrowseComp-Plus, the dual-system agents achieve a 3.7-4.7 times speedup in active search time while improving answer accuracy from 45% to up to 54%.

发表机构

• Huawei Technologies Co., Ltd.（华为技术有限公司）
• Leiden University（莱顿大学）

机构由 AI 辅助整理，请以论文原文为准。