FlowingFlowing
← Back to all development cases
Development case · blind model judgment pending independent human review; not a formal Benchmark conclusion.
AutocompleteEvidence version preferred3 / 162 · ce22995352c0cdc1

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

Quantitative biology · 2011.10873v1

FLOWING EVIDENCE BENCHMARK

How do we tell whether Evidence helps?

At the same writing position, with the same model and task, how does supplying retrieved paper passages change the first output? We compare matched versions and retain ties, unusable outputs and incomplete reviews.

SAME DRAFT · EVIDENCE ON OR OFF

01 · FIXED WRITING POSITION

MANUSCRIPT

Same writing position ▌

Matched autocomplete at the same draft position

02 · TWO MATCHED INPUTS

SHARED BY BOTH

Manuscript context, model, task and prompt

A · With Evidence

Retrieved paper passages supplied

B · Without Evidence

No retrieved passages supplied

LLM

Same model and version

A → first output

B → first output

Blind judge agent

First outputs are anonymized as X and Y

Continuation review: accuracy, fit to the writing task and usability
Returns: X preferred / tie / Y preferred / both unusable

Order check: X / Y → Y / X

HOW DOES BLIND JUDGING WORK?

① Anonymize both outputs
The judge sees the same draft and both first outputs without knowing which received Evidence.

② Compare and swap order
The judge applies task-specific criteria in X/Y and then Y/X order.

③ Review disagreements
A third pass resolves disagreements. Incomplete reviews remain in the denominator.

How are autocomplete positions stratified?

Before seeing generation outcomes, we check whether a retrieved passage contains a specific proposition that directly supports the next writing move. Those positions appear in the left opportunity group; the rest are ordinary positions on the right. We select a balanced sample from admitted papers in each field. The 50/50 split is experimental, not a measure of how often either type occurs in writing.

AUTOCOMPLETE · 110

Biology, statistics and astrophysics

55 positions on each side; statistics uses the ten-paper rerun.

AUTOCOMPLETE · 52

Psychology and climate science

26 positions on each side; psychology includes nine papers and climate science four.

How are the table percentages calculated?

Across five fields, 51 of 81 left-column positions preferred the Evidence version. Ties, pairs where both versions were unusable and incomplete reviews remain in the denominator.

51Evidence version preferred
÷
81All positions in this group
=
63%Evidence preference in this group

Source contribution is a separate review: 16 of 22 Evidence wins entered into source review directly used retrieved papers; another 29 wins await review.

These are development-stage model judgments pending independent human review. They are not formal Benchmark conclusions and do not, on their own, establish causality.

Manuscript writing position

Text excerpt · not a PDF page

Research manuscript · excerpt

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

5 Discussion > 5.2 Level of detail

…sociating rule of behavior with a specific hierarchy is possible once we have sufficient familiarity with the biology and predictions of basic models. This can help to bridge hierarchies and avoid the fog of complexity. Mathematical methods such as dynamical systems, PDEs, and bifurcation analysis can help to find and account for emergent properties and universal principles in such basic models [ 48 ] . This is one of the strengths of the mathematical tools. A weakness is that these methods currently work well for small systems of differential equations, but not for large and complex systems.

…perimental observations of small cell groups provide good opportunities for understanding how to bridge from single to collective behavior. Modeling can then also explore the advantages of group migration, chemotaxis [ 18 , 103 ] , durotaxis [ 165 ] , etc., in larger groups. For a large enough number of cells, condensing the details into simpler rules becomes expedient. For example, while polarity is represented by PDEs and patterns inside a single cell, it can then be simplified, depicted by a direction vector [ 25 , 15 ] in place of a full internal gradient of Rho or Rac for multiple cells.

↓ Next paragraph was generated here

The highlighted text is near the writing position. This is a layout of the evaluation input, not a PDF screenshot.

A matched comparison at the same writing positionJudgment: Evidence version preferred

Both arms use the same model, task and manuscript context. Only the retrieved passages differ. The first recorded output from each arm is shown below.

Evidence

Same model and position · retrieved passages supplied · first output

Preferred by blind review

The level of detail required in a model therefore depends on the question being asked rather than on the size of the system alone. A model that resolves intracellular polarity dynamics may be necessary to explain how a single cell reorients, whereas a direction-vector representation may suffice to capture the collective behavior of a large group.

Passages supplied to the Evidence version

Spontaneous Spatiotemporal Ordering of Shape Oscillations Enhances Cell Migration

Here we demonstrate, using numerical simulations, that the interplay of directed motion, shape oscillations, and excluded volume enables cells to locally “synchronize” their motion and thus enhance collective migration. Our model captures elongation and contraction of crawling ameboid cells controll…

Read full passage excerpt

Here we demonstrate, using numerical simulations, that the interplay of directed motion, shape oscillations, and excluded volume enables cells to locally “synchronize” their motion and thus enhance collective migration. Our model captures elongation and contraction of crawling ameboid cells controlled by an internal clock with a fixed period, mimicking the internal cycle of biological cells. We show that shape oscillations are crucial for local rearrangements that induce ordering of internal clocks between neighboring cells even in the absence of signaling and regularization. Our findings reveal a novel, purely physical mechanism through which the internal dynamics of cells influences their collective behavior, which is distinct from well known mechanisms like chemotaxis, cell division, and cell-cell adhesion.

No Evidence

Same model and position · no retrieved passages · first output

Not preferred in this comparison

A related limitation concerns the level of detail at which cell-cell interactions are represented. Detailed models of contact mechanics and receptor-mediated adhesion capture the behavior of small groups, but the computational cost of resolving these interactions grows rapidly with cell number, which motivates the simplified interaction rules used in larger-scale collective models.

Blind model judgment

The blinded model review preferred the Evidence version. Its original rationale appears below.

Read the original model judgment

Output A builds coherently from the draft's contrast between detailed PDE-based single-cell models and simplified direction-vector representations for multiple cells, extending this to the general principle that model detail depends on the research question. This is grounded in the draft's explicit discussion of simplifying polarity from PDEs to direction vectors [25, 15] for multiple cells. Output B introduces an unsupported claim about 'contact mechanics and receptor-mediated adhesion' and their 'computational cost' growing with cell number — neither contact mechanics, receptor-mediated adhesion, nor computational cost scaling are mentioned in the draft or source. The source discusses shape oscillations and excluded volume as a physical mechanism for collective migration, not contact mechanics or adhesion costs. Output B's claim about 'simplified interaction rules used in larger-scale collective models' also goes beyond the draft's specific point about direction vectors replacing PDEs for polarity.

The passages are input material; their presence does not mean an output used them or that they support every claim. Draft and source passages are excerpted; no simulated scores or PDF appear here.