4B 设备上的深度研究：曝光界限忠实度、检索界限覆盖率
On-Device Deep Research at 4B: Exposure Bounds Faithfulness, Retrieval Bounds Coverage
浏览论文内容
中文总结 AI 辅助
该研究聚焦 4B 设备上深度研究，区分引用主张忠实度与可信覆盖率两个量，通过交叉对比来源字符数和质量，发现曝光决定忠实度，检索决定覆盖率，提出先提高曝光再调控检索召回率的实际方法。
中文摘要 AI 辅助
设备上的研究代理在个人笔记本电脑上搜索语料库、阅读来源并撰写引用摘要。对于可部署的小模型，其引用是否忠实以及成本如何尚无衡量。本研究在 24GB 笔记本电脑上固定一个 4B 生成器，探究使其引用忠实的因素。区分了通常合并报告的两个量：引用主张忠实度，即引用来源是否支持主张；可信覆盖率，即代理是否引用了正确来源。研究将生成器所见每个来源的字符数（分别为 400 和 1500 字符）与所提供来源的质量（金论文与检索到的论文）进行交叉对比。发现曝光决定忠实度，更多曝光可提升忠实度，且两种设置下忠实度趋于一致，曝光提升对第二位独立评判者也稳健；检索决定覆盖率，在任何曝光下，检索来源的可信覆盖率接近 0.22，因为召回率保持在 0.40 左右，曝光无法决定引用哪些来源，并得出实际方法为先廉价提高每个来源的曝光，再将检索召回率作为唯一调控手段。
英文摘要
On-device research agents search a corpus, read sources, and write a cited brief on a personal laptop. Whether their citations are faithful, and at what cost, is unmeasured for a deployable small model. This study fixes one 4B generator on a 24 GB laptop and asks what makes its citations faithful. It separates two quantities usually reported as one number. Cited claim faithfulness asks whether the cited source supports the claim. Trustworthy coverage asks whether the agent also cites the right sources. The study crosses how much of each source the generator sees, 400 against 1500 characters, with the quality of the sources supplied, gold papers against retrieved papers. Two levers fall out, and they act on different outcomes. Exposure sets faithfulness. More of each source lifts faithfulness from 0.45 to 0.58 on retrieved sources and from 0.37 to 0.58 on gold sources, and the two settings converge, so faithfulness is bound by exposure, not by whether the source is correct. The exposure lift is robust to a second, independent judge; the exact convergence is tight under the primary judge and only approximate under the second. Retrieval sets coverage. Trustworthy coverage stays near 0.22 on retrieved sources at any exposure, because recall is held near 0.40, so exposure cannot fix which sources are cited. The extra exposure costs about 235 output tokens. The practical recipe is to raise per source exposure first, cheaply, and then treat retrieval recall as the only remaining lever.
