Evaluation

NTv2-250Mᵃ on the RVDB long-sequence dataset, 16 classes

Published comparison; transcribed, not reproduced.

Evaluation results

1 evaluation · 6 results. Different protocols are not a single leaderboard.

Protocol: RVDB long-sequence viral classification, 16 classes (Bůžková et al. 2026 Table 10) · Dataset: RVDB C-RVDB v31.0 long-sequence dataset, 16 classes

Sorted by Accuracy (16-class long-sequence RVDB dataset, chunk-scored) (higher is better). The best value in each column is highlighted. Decimals are rounded for display; each value links to the printed value and its source.

Tested configuration	Accuracy (16-class long-sequence RVDB dataset, chunk-scored)	Balanced accuracy (16-class long-sequence RVDB dataset, chunk-scored)	Macro F1 (16-class long-sequence RVDB dataset, chunk-scored)	MCC (16-class long-sequence RVDB dataset, chunk-scored)	Precision (16-class long-sequence RVDB dataset, chunk-scored, macro-averaged over classes)	Recall (16-class long-sequence RVDB dataset, chunk-scored, macro-averaged over classes)
NTv2-250M, frozen embeddings, pooling not stated, 1-NN, long-sequence run (Bůžková et al. 2026)Evaluation	0.831	0.844	0.842	0.818	0.844	0.844

All 6 result rows with coverage, uncertainty and sources

Filter evaluations

Tested model or methodProtocolDatasetEvaluation setupMetricEvidence origin

Applied filters: All linked evaluations

Exact evaluated configurations and original reported results
Tested configuration	Protocol and dataset	Finding	Evidence and details
Configuration: NTv2-250M, frozen embeddings, pooling not stated, 1-NN, long-sequence run (Bůžková et al. 2026)	Protocol: RVDB long-sequence viral classification, 16 classes (Bůžková et al. 2026 Table 10) Dataset: RVDB C-RVDB v31.0 long-sequence dataset, 16 classes	0.831 Accuracy (16-class long-sequence RVDB dataset, chunk-scored) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source NTv2-250Mᵃ on the RVDB long-sequence dataset, 16 classes gfm2-20261011-buzkova2026-protocol-long Aggregation: Not reported Analyzing Genomic Foundation Models for Viral Sequence Identification · Table 10, row 'NTv2-250Mᵃ', column 'Accuracy'
Configuration: NTv2-250M, frozen embeddings, pooling not stated, 1-NN, long-sequence run (Bůžková et al. 2026)	Protocol: RVDB long-sequence viral classification, 16 classes (Bůžková et al. 2026 Table 10) Dataset: RVDB C-RVDB v31.0 long-sequence dataset, 16 classes	0.844 Balanced accuracy (16-class long-sequence RVDB dataset, chunk-scored) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source NTv2-250Mᵃ on the RVDB long-sequence dataset, 16 classes gfm2-20261011-buzkova2026-protocol-long Aggregation: Not reported Analyzing Genomic Foundation Models for Viral Sequence Identification · Table 10, row 'NTv2-250Mᵃ', column 'Bal. Acc.'
Configuration: NTv2-250M, frozen embeddings, pooling not stated, 1-NN, long-sequence run (Bůžková et al. 2026)	Protocol: RVDB long-sequence viral classification, 16 classes (Bůžková et al. 2026 Table 10) Dataset: RVDB C-RVDB v31.0 long-sequence dataset, 16 classes	0.842 Macro F1 (16-class long-sequence RVDB dataset, chunk-scored) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source NTv2-250Mᵃ on the RVDB long-sequence dataset, 16 classes gfm2-20261011-buzkova2026-protocol-long Aggregation: Not reported Analyzing Genomic Foundation Models for Viral Sequence Identification · Table 10, row 'NTv2-250Mᵃ', column 'F1'
Configuration: NTv2-250M, frozen embeddings, pooling not stated, 1-NN, long-sequence run (Bůžková et al. 2026)	Protocol: RVDB long-sequence viral classification, 16 classes (Bůžková et al. 2026 Table 10) Dataset: RVDB C-RVDB v31.0 long-sequence dataset, 16 classes	0.818 MCC (16-class long-sequence RVDB dataset, chunk-scored) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source NTv2-250Mᵃ on the RVDB long-sequence dataset, 16 classes gfm2-20261011-buzkova2026-protocol-long Aggregation: Not reported Analyzing Genomic Foundation Models for Viral Sequence Identification · Table 10, row 'NTv2-250Mᵃ', column 'MCC'
Configuration: NTv2-250M, frozen embeddings, pooling not stated, 1-NN, long-sequence run (Bůžková et al. 2026)	Protocol: RVDB long-sequence viral classification, 16 classes (Bůžková et al. 2026 Table 10) Dataset: RVDB C-RVDB v31.0 long-sequence dataset, 16 classes	0.844 Precision (16-class long-sequence RVDB dataset, chunk-scored, macro-averaged over classes) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source NTv2-250Mᵃ on the RVDB long-sequence dataset, 16 classes gfm2-20261011-buzkova2026-protocol-long Aggregation: Not reported Analyzing Genomic Foundation Models for Viral Sequence Identification · Table 10, row 'NTv2-250Mᵃ', column 'Precision'
Configuration: NTv2-250M, frozen embeddings, pooling not stated, 1-NN, long-sequence run (Bůžková et al. 2026)	Protocol: RVDB long-sequence viral classification, 16 classes (Bůžková et al. 2026 Table 10) Dataset: RVDB C-RVDB v31.0 long-sequence dataset, 16 classes	0.844 Recall (16-class long-sequence RVDB dataset, chunk-scored, macro-averaged over classes) fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source NTv2-250Mᵃ on the RVDB long-sequence dataset, 16 classes gfm2-20261011-buzkova2026-protocol-long Aggregation: Not reported Analyzing Genomic Foundation Models for Viral Sequence Identification · Table 10, row 'NTv2-250Mᵃ', column 'Recall'

Source checking is not independent reproduction. Release 2026-11-02-a7f03c9e55a5.

Research readiness

0 of 4 readiness checks met. These checks assess whether the evidence supports a reproducible investigation; a source-checked score alone does not meet them.

Readiness checks, gaps and artifacts

Release 2026-11-02-a7f03c9e55a5 · Evidence verified: Not verified

Evidence incomplete

Replay metrics

Exact outcomes, predictions, identifiers and evaluator are connected.

Missing or unresolved evidence

No verified artifact manifest is linked to this exact record.
File checksums match the recorded files: not yet verified
Predictions are matched to the right samples: not yet verified
Score meaning and direction are confirmed: not yet verified
Metrics are recomputed from the saved predictions: not yet verified

Verified: Not verified

Evidence incomplete

Investigate discrepancies

Replay evidence includes annotations and an assessment of dependence. Unknown independence permits descriptive analysis only.

Missing or unresolved evidence

No verified artifact manifest is linked to this exact record.
File checksums match the recorded files: not yet verified
Predictions are matched to the right samples: not yet verified
Score meaning and direction are confirmed: not yet verified
Metrics are recomputed from the saved predictions: not yet verified
Sample annotations are recorded: not yet verified
Dependence between samples is assessed: not yet verified

Verified: Not verified

Evidence incomplete

Run locally

A pinned recipe describes the inputs, environment and resource requirements.

Missing or unresolved evidence

No verified artifact manifest is linked to this exact record.
File checksums match the recorded files: not yet verified
Predictions are matched to the right samples: not yet verified
Score meaning and direction are confirmed: not yet verified
A pinned run recipe exists: not yet verified
Compute requirements are estimated: not yet verified

Verified: Not verified

Evidence incomplete

Validate independently

Separate data and exposure records support an independent test.

Missing or unresolved evidence

No verified artifact manifest is linked to this exact record.
File checksums match the recorded files: not yet verified
Predictions are matched to the right samples: not yet verified
Score meaning and direction are confirmed: not yet verified
Independent validation data exist: not yet verified
Overlap with training data is checked: not yet verified

Verified: Not verified

Readiness describes the evidence in this release. Availability on your computer is checked separately when an investigation runs. Existing data exposure can prevent independent validation even when files are available.

Artifacts and reproduction

No verified artifact manifest is connected to this record yet. The gaps above identify what is needed before analysis can begin.

Evaluation procedure

RVDB long-sequence viral classification, 16 classes (Bůžková et al. 2026 Table 10)

Configuration: NTv2-250M, frozen embeddings, pooling not stated, 1-NN, long-sequence run (Bůžková et al. 2026)
Protocol: RVDB long-sequence viral classification, 16 classes (Bůžková et al. 2026 Table 10)
Dataset: RVDB C-RVDB v31.0 long-sequence dataset, 16 classes

origin: Independent external evaluation
configuration: Primary source as retrieved 2026-10-11
protocol id: gfm2-20261011-buzkova2026-protocol-long
dataset version: C-RVDB v31.0 long-sequence dataset, 16 classes
split: About 4 indexed and about 16 queried sequences per class
population: 256 test sequences
inputs: Whole viral genome, 84,587 to 409,110 bp, scored over overlapping chunks
adaptation: None: frozen weights, pooling not stated, 1-NN over the indexed training embeddings
metric implementation: Accuracy, balanced accuracy, MCC and macro-averaged F1, precision and recall
aggregation: Over chunks
budget: Not reported

Metadata review: source checked. Unreported conditions prevent automatic comparisons.

Reproduction

Split: About 4 indexed and about 16 queried sequences per class
Adaptation: None: frozen weights, pooling not stated, 1-NN over the indexed training embeddings
Scoring implementation: Accuracy, balanced accuracy, MCC and macro-averaged F1, precision and recall

No execution recipe has been verified for this exact configuration and evaluation. A benchmark's general instructions may use different inputs, splits or model settings.

Reproducing this published result requires matching its model configuration, data, split and scorer. Source checking or a successful smoke test does not establish score reproduction.

Evidence

Source checking verifies the cited claim or transcription. It does not establish independent reproduction.

Evidence table

Inspect claims, sources and review details

Trace each statement to its source and review. A context-only reference supports the record generally; it does not verify an individual field. Source checking does not reproduce an experiment.

One row per statement and cited source. Multiple citations are not independent evaluations. Shared locators are labelled explicitly.

19 evidence rows matching the loaded filters

Claims, original sources and review scope · Release 2026-11-02-a7f03c9e55a5
Property and statement	Original source and location	Review and provenance
Comparison: adaptation None: frozen weights, pooling not stated, 1-NN over the indexed training embeddings Context-only references	Analyzing Genomic Foundation Models for Viral Sequence Identification Original source ↗ Table 10, row 'NTv2-250Mᵃ' Version: bioRxiv v1, posted 2026-09-28 Retrieved: 2026-10-11T00:48:35Z	not individually reviewed No individual claim review recorded independent paper Audit details Field: `attributes.comparison.adaptation` Source artifact SHA-256: `9a589a2088f8466502b1035228db3e594065e984e5755c25d907471225f75c19` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Comparison: aggregation Over chunks Context-only references	Analyzing Genomic Foundation Models for Viral Sequence Identification Original source ↗ Table 10, row 'NTv2-250Mᵃ' Version: bioRxiv v1, posted 2026-09-28 Retrieved: 2026-10-11T00:48:35Z	not individually reviewed No individual claim review recorded independent paper Audit details Field: `attributes.comparison.aggregation` Source artifact SHA-256: `9a589a2088f8466502b1035228db3e594065e984e5755c25d907471225f75c19` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Comparison: budget Not reported Context-only references	Analyzing Genomic Foundation Models for Viral Sequence Identification Original source ↗ Table 10, row 'NTv2-250Mᵃ' Version: bioRxiv v1, posted 2026-09-28 Retrieved: 2026-10-11T00:48:35Z	missing or unspecified No individual claim review recorded independent paper Audit details Field: `attributes.comparison.budget` Source artifact SHA-256: `9a589a2088f8466502b1035228db3e594065e984e5755c25d907471225f75c19` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Comparison: dataset version C-RVDB v31.0 long-sequence dataset, 16 classes Context-only references	Analyzing Genomic Foundation Models for Viral Sequence Identification Original source ↗ Table 10, row 'NTv2-250Mᵃ' Version: bioRxiv v1, posted 2026-09-28 Retrieved: 2026-10-11T00:48:35Z	not individually reviewed No individual claim review recorded independent paper Audit details Field: `attributes.comparison.dataset_version` Source artifact SHA-256: `9a589a2088f8466502b1035228db3e594065e984e5755c25d907471225f75c19` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Comparison: inputs Whole viral genome, 84,587 to 409,110 bp, scored over overlapping chunks Context-only references	Analyzing Genomic Foundation Models for Viral Sequence Identification Original source ↗ Table 10, row 'NTv2-250Mᵃ' Version: bioRxiv v1, posted 2026-09-28 Retrieved: 2026-10-11T00:48:35Z	not individually reviewed No individual claim review recorded independent paper Audit details Field: `attributes.comparison.inputs` Source artifact SHA-256: `9a589a2088f8466502b1035228db3e594065e984e5755c25d907471225f75c19` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Comparison: metric implementation Accuracy, balanced accuracy, MCC and macro-averaged F1, precision and recall Context-only references	Analyzing Genomic Foundation Models for Viral Sequence Identification Original source ↗ Table 10, row 'NTv2-250Mᵃ' Version: bioRxiv v1, posted 2026-09-28 Retrieved: 2026-10-11T00:48:35Z	not individually reviewed No individual claim review recorded independent paper Audit details Field: `attributes.comparison.metric_implementation` Source artifact SHA-256: `9a589a2088f8466502b1035228db3e594065e984e5755c25d907471225f75c19` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Comparison: population 256 test sequences Context-only references	Analyzing Genomic Foundation Models for Viral Sequence Identification Original source ↗ Table 10, row 'NTv2-250Mᵃ' Version: bioRxiv v1, posted 2026-09-28 Retrieved: 2026-10-11T00:48:35Z	not individually reviewed No individual claim review recorded independent paper Audit details Field: `attributes.comparison.population` Source artifact SHA-256: `9a589a2088f8466502b1035228db3e594065e984e5755c25d907471225f75c19` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Comparison: protocol id gfm2-20261011-buzkova2026-protocol-long Context-only references	Analyzing Genomic Foundation Models for Viral Sequence Identification Original source ↗ Table 10, row 'NTv2-250Mᵃ' Version: bioRxiv v1, posted 2026-09-28 Retrieved: 2026-10-11T00:48:35Z	not individually reviewed No individual claim review recorded independent paper Audit details Field: `attributes.comparison.protocol_id` Source artifact SHA-256: `9a589a2088f8466502b1035228db3e594065e984e5755c25d907471225f75c19` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Comparison: split About 4 indexed and about 16 queried sequences per class Context-only references	Analyzing Genomic Foundation Models for Viral Sequence Identification Original source ↗ Table 10, row 'NTv2-250Mᵃ' Version: bioRxiv v1, posted 2026-09-28 Retrieved: 2026-10-11T00:48:35Z	not individually reviewed No individual claim review recorded independent paper Audit details Field: `attributes.comparison.split` Source artifact SHA-256: `9a589a2088f8466502b1035228db3e594065e984e5755c25d907471225f75c19` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Limitations 3 values Independent evaluation; the authors develop none of the methods compared. Scored over overlapping chunks rather than whole sequences (Table 10 footnote a), so not the same unit as the other rows. Section 2.5 says chunk embeddings were pooled to a sequence-level representation, but the Table 10 footnote and Section 4 (Methodological considerations) say these rows were scored over chunks; the scored unit is recorded as the footnote states. Context-only references	Analyzing Genomic Foundation Models for Viral Sequence Identification Original source ↗ Table 10, row 'NTv2-250Mᵃ' Version: bioRxiv v1, posted 2026-09-28 Retrieved: 2026-10-11T00:48:35Z	not individually reviewed No individual claim review recorded independent paper Audit details Field: `attributes.limitations` Source artifact SHA-256: `9a589a2088f8466502b1035228db3e594065e984e5755c25d907471225f75c19` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact

How to read the evidence

Sources and history

Release 2026-11-02-a7f03c9e55a5 · Record review: source checked

1 source record and release history

Analyzing Genomic Foundation Models for Viral Sequence Identification · Original source · bioRxiv v1, posted 2026-09-28

Download this release (gzip)

Technical metadata and extraction receipts

Stable ID: gfm2-20261011-buzkova2026-eval-t10-ntv2-250m

areas: dna-genomes
contexts: research
origin: independent_paper
protocol: gfm2-20261011-buzkova2026-protocol-long
version: Primary source as retrieved 2026-10-11
comparison: protocol id: gfm2-20261011-buzkova2026-protocol-long; dataset version: C-RVDB v31.0 long-sequence dataset, 16 classes; split: About 4 indexed and about 16 queried sequences per class; population: 256 test sequences; inputs: Whole viral genome, 84,587 to 409,110 bp, scored over overlapping chunks; adaptation: None: frozen weights, pooling not stated, 1-NN over the indexed training embeddings; metric implementation: Accuracy, balanced accuracy, MCC and macro-averaged F1, precision and recall; aggregation: Over chunks; budget: Not reported
source locator: Table 10, row 'NTv2-250Mᵃ'
limitations: Independent evaluation; the authors develop none of the methods compared.; Scored over overlapping chunks rather than whole sequences (Table 10 footnote a), so not the same unit as the other rows.; Section 2.5 says chunk embeddings were pooled to a sequence-level representation, but the Table 10 footnote and Section 4 (Methodological considerations) say these rows were scored over chunks; the scored unit is recorded as the footnote states.

Related records

system: NTv2-250M, frozen embeddings, pooling not stated, 1-NN, long-sequence run (Bůžková et al. 2026)
assessment: RVDB long-sequence viral classification, 16 classes (Bůžková et al. 2026 Table 10)
data: RVDB C-RVDB v31.0 long-sequence dataset, 16 classes
evaluation: t10-NTv2-250M-Accuracy accuracy
evaluation: t10-NTv2-250M-Bal. Acc. balanced-accuracy
evaluation: t10-NTv2-250M-F1 macro-f1
evaluation: t10-NTv2-250M-MCC matthews-correlation-coefficient
evaluation: t10-NTv2-250M-Precision precision
evaluation: t10-NTv2-250M-Recall recall

Suggest a correction