FlowingFlowing
← Back to all development cases
Development case · blind model judgment pending independent human review; not a formal Benchmark conclusion.
AutocompleteEvidence version preferred34 / 162 · cf64b9ac8ea7c26b

A Review of BioTree Construction in the Context of Information Fusion: Priors, Methods, Applications and Trends

Quantitative biology · 2410.04815v2

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 BioTree Construction in the Context of Information Fusion: Priors, Methods, Applications and Trends

6 Classical BioTree Construction Methods > 6.5 Limitations of Traditional BioTree Construction Methods

Similarity matrix-based methods utilize a cell-to-cell similarity matrix to analyze relationships between cells and construct lineage trees based on these similarities. SoptSC [ 298 ] builds a lineage tree by performing clustering and lineage inference using cell-cell relationships derived from a similarity matrix, effectively capturing the hierarchical differentiation paths in a tree structure.

…ccount for noise and stochasticity in gene expression profiles while constructing lineage trees. cellTree [ 65 ] models the gene expression data using a probabilistic framework to construct a tree-like structure that outlines hierarchical differentiation, explicitly representing cell lineages as branches of a tree. CALISTA [ 214 ] integrates clustering, lineage progression, transition gene identification, and pseudotime ordering into a unified framework, constructing lineage trees that represent the developmental trajectories of cells based on statistical modeling of gene expression patterns.

↓ 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

No continuation text was produced.

The model requested more evidence and did not continue.

Passages supplied to the Evidence version

Reconstructing probabilistic trees of cellular differentiation from single-cell RNA-seq data

Thanks to technological advances, we can now profile gene expression across thousands or millions of individual cells in parallel. This new type of data has led to the intriguing discovery that individual cell profiles can reflect the imprint of time or dynamic processes. However, synthesizing this…

Read full passage excerpt

Thanks to technological advances, we can now profile gene expression across thousands or millions of individual cells in parallel. This new type of data has led to the intriguing discovery that individual cell profiles can reflect the imprint of time or dynamic processes. However, synthesizing this information to reconstruct dynamic biological phenomena from data that are noisy, heterogenous, and sparse—and from processes that may unfold asynchronously—poses a complex computational and statistical challenge. Here, we develop a full generative model for probabilistically reconstructing trees of cellular differentiation from single-cell RNA-seq data. Specifically, we extend the framework of the classical Dirichlet diffusion tree to simultaneously infer branch topology and latent cell states along continuous trajectories over the full tree.

No Evidence

Same model and position · no retrieved passages · first output

Not preferred in this comparison

Despite these capabilities, similarity matrix-based methods face several inherent limitations. Their performance depends heavily on the accuracy of the estimated cell-to-cell similarity matrix, which is sensitive to noise, dropout events, and the choice of distance metric.

Blind model judgment

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

Read the original model judgment

Output A adds material factual claims that are not supported by the draft or supplied sources. It states that similarity matrix-based methods 'face several inherent limitations,' that their performance 'depends heavily on the accuracy of the estimated cell-to-cell similarity matrix,' and that this matrix 'is sensitive to noise, dropout events, and the choice of distance metric.' None of these claims appear in the draft text, which only describes methods and their approaches without evaluating limitations. The single supplied source discusses cellTree's probabilistic framework and general challenges of single-cell RNA-seq data (noisy, heterogeneous, sparse, asynchronous), but does not mention similarity matrix-based methods, their limitations, dropout events, or distance metric sensitivity. The source's general discussion of noise and sparsity in single-cell data does not support specific limitations of similarity matrix-based methods. Output B correctly returns an empty result with 'needs_evidence' action, acknowledging that no supplied source directly supports a completion for this blank paragraph.

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.