FlowingFlowing
← Back to all development cases
Development case · blind model judgment pending independent human review; not a formal Benchmark conclusion.
AutocompleteNo-Evidence version preferred24 / 162 · 6c425bc3c01280e6

Deciphering genomic codes using advanced NLP techniques: a scoping review

Quantitative biology · 2411.16084v1

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

Deciphering genomic codes using advanced NLP techniques: a scoping review

4 Discussion

One limitation of this study is the constrained scope of the literature review sources, which primarily includes human genome data and only limited exploration of bacterial, viral, and other non-human DNA. Moreover, the study predominantly focuses on cancer when it comes to disease analysis, giving relatively less attention to other disease domains that involve complex DNA interactions, such as neurodegenerative diseases, autoimmune diseases, and genetic disorders. These areas also offer rich opportunities for genomic research and could benefit from applying NLP techniques.

… Including clinical data can enhance model predictions by grounding them in real-world patient information, thereby improving clinical relevance and enabling personalized insights. This comprehensive approach can provide a more comprehensive understanding of the regulatory mechanisms governing gene expression and the interplay between different molecular layers. It can also enable models to generate and validate more accurate and biologically meaningful predictions, thereby increasing physicians’ confidence in NLP-generated results and promoting the widespread application of NLP-based models.

↓ 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: No-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

Not preferred in this comparison

No continuation text was produced.

Model action：complete

Passages supplied to the Evidence version

Mapping Dynamic Histone Acetylation Patterns to Gene Expression in Nanog-depleted Murine Embryonic Stem Cells

Over the following five days they measured changes on four different molecular levels: histone acetylation (H3K9,14ac), chromatin-bound RNA polymerase II, messenger RNA (mRNA) expression and nuclear protein abundance. This data set provides a rich resource to untangle the complexity of the multi-lay…

Read full passage excerpt

Over the following five days they measured changes on four different molecular levels: histone acetylation (H3K9,14ac), chromatin-bound RNA polymerase II, messenger RNA (mRNA) expression and nuclear protein abundance. This data set provides a rich resource to untangle the complexity of the multi-layer regulatory mechanism responsible for stem cell fate. Lu et al anchored their analyses on changes in nuclear protein expression and found that many lacked concordant changes in mRNA expression, pointing to important roles for translational and post-translational regulation of ESC fate. Here, we complement theses analyses with an in-depth study of the relation between histone acetylation and gene expression in the same data set.

No Evidence

Same model and position · no retrieved passages · first output

Preferred by blind review

Beyond expanding the data modalities, the interpretability of NLP-derived genomic predictions remains a critical determinant of clinical adoption. Models that cannot expose the sequence features or regulatory logic underlying their outputs are unlikely to earn the trust of clinicians or regulators, even when their predictive accuracy is high.

Blind model judgment

The blinded model review preferred the no-Evidence version. Its original rationale appears below.

Read the original model judgment

Output A provides a coherent, relevant academic paragraph that logically extends the draft's discussion. The draft moves from limitations (narrow scope, cancer focus) to opportunities (including clinical data, multi-layer understanding, clinical relevance). Output A continues this trajectory by addressing interpretability as another critical factor for clinical adoption—directly connecting to the draft's closing themes of 'physicians' confidence' and 'widespread application.' The paragraph is self-contained, requires no editing, and preserves the manuscript's scope without adding unsupported factual claims. Output B is empty, which violates the task requirement to produce one coherent academic paragraph when the task clearly needs 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.