FlowingFlowing
← Back to all development cases
Development case · blind model judgment pending independent human review; not a formal Benchmark conclusion.
AutocompleteBoth unusable28 / 162 · f1f3ac87fc1a611d

A Review of Link Prediction Applications in Network Biology

Quantitative biology · 2312.01275v1

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

A Review of Link Prediction Applications in Network Biology

7. Conclusion

…eep machine learning-based dynamic embedding techniques ( koutrouli2020guide ) as well as higher-order analysis that takes into account network substructures (or motifs ) ( roy2020motifs ) . There have been a few efforts to predict links based on the motif participation of nodes ( abuoda2020link ) or the presence or absence of motifs ( roy2023inferring ) . A recent embedding algorithm combines motifs with deep learning approaches to predict associations ( wang2020model ) . Finally, others have combined network reconstruction to meet link prediction goals ( wang2017kernel ; wu2019enhancing ) .

…ge graph may possess some common and some rare associations. This class imbalance is likely to lead to biased predictions, (b) Semantic heterogeneity. Biological entities are often described using diverse ontologies and terminologies. Thus, link prediction models must handle the semantic heterogeneity by mapping or aligning different entity types and relationships, and (c) Incorporating multi-omics data. Knowledge graphs include multi-omics data, such as genomics, proteomics, and metabolomics. LP models must assimilate data from these diverse sources to make biologically relevant predictions.

↓ 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: Both unusable

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

Judged unusable

No continuation text was produced.

Model action：complete

Passages supplied to the Evidence version

Fusing heterogeneous data sets

These different sources of measurements not only provide us with the opportunity of a deeper understanding of the studied system, but they also introduce some new statistical challenges. All these challenges are related to the heterogeneity of the data sets. The first type of heterogeneity is the ty…

Read full passage excerpt

These different sources of measurements not only provide us with the opportunity of a deeper understanding of the studied system, but they also introduce some new statistical challenges. All these challenges are related to the heterogeneity of the data sets. The first type of heterogeneity is the type of data , such as metabolomics, proteomics and RNAseq data in genomics. These different omics data reflect the properties of the studied biological system from different perspectives. The second type of heterogeneity is the type of scale , which indicates the measurements are obtained at different scales, such as binary, ordinal, interval and ratio-scaled variables. Within this thesis, various data fusion approaches are developed to tackle either one or two types of heterogeneity that exist in multiple data sets.

No Evidence

Same model and position · no retrieved passages · first output

Judged unusable

No continuation text was produced.

Model action：complete

Blind model judgment

The blind model judge marked both outputs unusable; this is counted separately from a tie.

Read the original model judgment

Both outputs A and B are empty strings with action 'complete'. The task requires continuing the manuscript with one coherent academic paragraph. The draft ends with three challenges in biological knowledge graph link prediction: (a) class imbalance, (b) semantic heterogeneity, and (c) incorporating multi-omics data. The source provided discusses heterogeneity in multi-omics data fusion (metabolomics, proteomics, RNAseq), which directly relates to challenge (c). A usable continuation would need to elaborate on these challenges or transition to discussing solutions, grounded in the available source material. Empty outputs fail to fulfill the task requirement of producing content.

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.