FlowingFlowing
← 返回全部开发案例
开发阶段案例 · 模型盲评尚待独立人工复核，不代表正式 Benchmark 结论。
自动补全无 Evidence 更优1 / 162 · 2c19f4cf93463f45

Bridging from single to collective cell migration: A review of models and links to experiments

定量生物学 · 2011.10873v1

FLOWING EVIDENCE BENCHMARK

怎么判断 Evidence 真的有帮助？

核心问题是：在同一写作位置、使用同一模型和任务时，提供检索文献片段会怎样改变首次输出？我们成对比较两种条件，保留持平、不可用和评审未完成的结果。

同一段原稿 · 比较是否提供 Evidence

01 · 固定写作位置

论文原稿

同一处写作点位 ▌

同一原稿位置的自动补全成对比较

02 · 构造两组输入

两组共用

原稿上下文、模型、任务和提示词

A · 提供 Evidence

额外提供检索文献片段

B · 不提供 Evidence

不提供检索文献片段

LLM

同一模型、同一版本

A → 首次输出

B → 首次输出

盲评 Agent

匿名标记两份首次输出为 X、Y

按续写质量判断：准确性、任务贴合度和可用性
输出：X 更优 / 持平 / Y 更优 / 两组均不可用

换序复评：X / Y → Y / X

盲评具体怎么判？

① 匿名两份输出
评审看到同一原稿和两份首次输出，但不知道哪份用了 Evidence。

② 比较并换序复评
根据当前任务的标准，按 X/Y、再按 Y/X 的顺序各评一次。

③ 复核分歧
两次结论不一致时再做第三次判定；未完成的评审也留在分母。

自动补全点位怎样分层？

在查看生成结果前，先核查检索片段是否含有能直接支撑下一步续写的具体命题；有则放在左侧机会组，否则放在右侧普通组。每个学科从可准入论文中均衡选取两组点位。50/50 是实验设计，不代表真实写作中两类点位各占一半。

AUTOCOMPLETE · 110

定量生物学、统计学、天体物理学

左侧 55 个、右侧 55 个点位；统计学采用 10 篇论文的复测结果。

AUTOCOMPLETE · 52

心理学与气候科学

左侧 26 个、右侧 26 个点位；心理学计入 9 篇，气候科学计入 4 篇。

表格里的百分比怎么算？

五学科共有 81 个左侧点位，其中 51 个点位的 Evidence 版本获评更优。持平、两组均不可用和评审未完成的点位仍计入分母。

51Evidence 版本更优
÷
81该层全部点位
=
63%该层 Evidence 获评更优的比例

来源贡献是另一项复核：已进入复核的 22 个 Evidence 获胜点位中，16 个确认直接使用了检索论文；另有 29 个胜出点位尚待复核。

当前是开发阶段的模型评审结果，尚未完成独立人工复核；这些数字不代表正式 Benchmark 结论，也不能单独证明因果关系。

原稿写作位置

原文摘录 · 非 PDF 页面

研究论文 · 原文片段

Bridging from single to collective cell migration: A review of models and links to experiments

1 Introduction

In this review paper, we summarize the literature on computational models for cell motility, from the biochemical networks that regulate it, to the behavior of 1 and many cells. We discuss the distinct approaches used at each level, and how models can build bridges between the different size scales. We find models at many different levels of biological detail, and discuss their relative contributions to our understanding of single and collective cell behavior. Finally, we indicate how models have been linked to biological experiments in this field.

Over several decades, there has been great progress in our understanding of cell motility. In the 1980s and 1990s, the basic machinery of eukaryotic cell motion and the role of the actin cytoskeleton were discovered and refined. Regulation of motility by intracellular signaling networks was then deciphered in the late 1990s and through the 2000s. We continue to discover links between cell signaling and cell shape and function, in both normal and diseased cells. Recent efforts aim to link single cell behavior to collective behavior of many cells and emergent dynamics of tissues.

↓ 此处生成下一段续写

黄色标出写作点位附近的原文；本页根据评测输入重排文本，不是原始 PDF 截图。

同一写作位置的成对对照盲评结果：无 Evidence 更优

两组使用同一模型、同一写作任务和原稿上下文，仅比较是否提供检索文献片段。下方展示评测记录中的首次输出。

Evidence

同一模型与写作位置 · 提供检索文献片段 · 首次输出

本次未获偏好

未输出续写正文。

模型表示需要更多证据，未生成续写。

提供给 Evidence 版本的文献片段

Collective motion of cells crawling on a substrate: roles of cell shape and contact inhibition

At the level of a single cell, it is well established that its motion is intricately linked to its shape. The shape of crawling cells is highly variable, depending on the type of cell, the substrate, as well as a result of the migration process itself [ 5 , 6 , 7 , 8 ] . When a cell starts moving, i…

展开完整文献摘录

At the level of a single cell, it is well established that its motion is intricately linked to its shape. The shape of crawling cells is highly variable, depending on the type of cell, the substrate, as well as a result of the migration process itself [ 5 , 6 , 7 , 8 ] . When a cell starts moving, its shape breaks symmetry [ 6 ] , whereas circular cells typically cannot move. While there is evidence that shape has a strong influence on scattering and can lead to clustering and collective directed motion of swimmers [ 9 , 10 ] , less is known about the role of cell shape in organizing collective crawling. It has been shown in simulations that inelastic collisions between crawling cells, e.g. due to deformation, can lead to coherent migration [ 11 , 12 , 13 , 14 ] , suggesting the importance of deformability for collective behavior.

无 Evidence

同一模型与写作位置 · 不提供检索文献片段 · 首次输出

盲评更优

This review focuses on the computational models that have been developed to describe cell motility at these different scales, and on how those models connect to experimental measurements. We organize the discussion from subcellular biochemistry through single-cell migration to collective and tissue-level behavior, highlighting where models bridge between scales and where gaps remain.

模型盲评结论

模型盲评判定“无 Evidence 更优”。下方保留评审原始理由（英文），供核对判断依据。

查看模型评审原始理由（英文）

Output A provides a coherent, usable continuation paragraph that fits the manuscript's structure and scope. It naturally bridges from the historical overview in the draft to the review's organizational framework, maintaining the paper's focus on computational models across scales. The content is supported by the draft's stated purpose (summarizing computational models for cell motility at different levels) and does not introduce unsupported factual claims. Output B returns an empty result with 'needs_evidence' action, which is unusable for an autocomplete task that clearly requires content continuation. The blank paragraph needs to be filled, and B fails to do so.

文献片段是输入材料；出现于此不代表输出使用了它，也不代表它能够支持全部主张。原文与检索片段经过截取；页面没有展示模拟分数或模拟 PDF。