Model type
Study-specific predictive method; this record is the paper-specific evaluated configuration.
Configuration
RNAret
RNAret learns RNA representations with a retention network instead of quadratic self-attention.
Sources
Retentive Network promotes efficient RNA language modeling of long sequences · Methods/Bidirectional encoder representation from Retentive Network (paragraph 1); Methods/Bidirectional encoder representation from Retentive Network (paragraph 7)How it worksEvaluated procedure (conceptual)
Conceptual input–method–output guide. Check the procedure text and linked evaluation for fitted components, additional inputs and exact settings.
Sources
Retentive Network promotes efficient RNA language modeling of long sequences · Methods/Bidirectional encoder representation from Retentive Network (paragraph 1); Methods/Bidirectional encoder representation from Retentive Network (paragraph 2)At a glance
Biological inputs
RNA nucleotide sequences
Sources
Retentive Network promotes efficient RNA language modeling of long sequences · Introduction (paragraph 2); Methods/RNA tokenization (paragraph 1)Outputs
RNA representations and downstream interaction, structure or classification predictions
Sources
Retentive Network promotes efficient RNA language modeling of long sequences · Methods/Fine-tuning strategies for downstream tasks (paragraph 1); Methods/RNA-RNA interaction prediction (paragraph 1)limited source coverage · Automated source review, 2026-09-16. All specifications and missing details
Evaluations and results
Release 2026-09-17-d277315f7d76 · 1 evaluation · 1 metric row. Different protocols are not a single leaderboard.
| Metric and finding | Coverage and uncertainty | Evidence |
|---|---|---|
| RNAret: miRNA-mRNA interaction prediction 5-mer RNAret classifier; 72/8/20 train/validation/test split Author-reported evaluation · Evaluation metadata: needs review | ||
| 0.9622 F1 Unit: fraction · Direction: unknown | Uncertainty: not reported in legacy extract Scored: Not reported · Eligible: Not reported | source checkedRetentive Network promotes efficient RNA language modeling of long sequences · Table 1, MirTarRAW / 5-mer RNAret row, F1 column Source checking is not independent reproduction. |
How it works
How the evaluated method works
Bidirectional retention blocks use multiscale retention, feed-forward layers, residual connections and layer normalisation; masked-language pretraining learns sequence features.
Sources
Retentive Network promotes efficient RNA language modeling of long sequences · Methods/Bidirectional encoder representation from Retentive Network (paragraph 1); Methods/Bidirectional encoder representation from Retentive Network (paragraph 2)What was evaluated
The linked evaluation record identifies RNAret: miRNA-mRNA interaction prediction. Its dataset, split, adaptation and evidence origin remain attached to the reported results.
Sources
Retentive Network promotes efficient RNA language modeling of long sequences · The named evaluation’s methods and comparison table; exact preserved evaluation IDs: evaluation-b2-rnaret-2026Strengths and limitations
Strengths and considerations
- The retention mechanism is designed to support longer sequences with linear scaling in the discussed processing regime.
Sources
Retentive Network promotes efficient RNA language modeling of long sequences · Methods/Bidirectional encoder representation from Retentive Network (paragraph 7); Methods/Bidirectional encoder representation from Retentive Network (paragraph 3)
Limitations and conditions
- Algorithmic scaling does not establish a universal maximum useful sequence length or equal accuracy at all lengths.
Sources
Retentive Network promotes efficient RNA language modeling of long sequences · Methods/Pretraining strategies (paragraph 3); Results/mRNA/lncRNA classification (paragraph 2)
Profile review details
Primary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction.
Stable record: reported-model-7234658bc9c828Specifications
Inputs, training, access and other details
Explanatory profile: limited source coverage · Automated source review, 2026-09-16. Review applies to the cited claims; unresolved fields are listed below. Numerical results retain their own review status.
| Property | Description and evidence |
|---|---|
| Model type | Study-specific predictive method; this record is the paper-specific evaluated configuration.SourcesRetentive Network promotes efficient RNA language modeling of long sequences · Methods/Bidirectional encoder representation from Retentive Network (paragraph 1); Methods/Bidirectional encoder representation from Retentive Network (paragraph 2) |
| Architecture / procedure | Bidirectional retention blocks use multiscale retention, feed-forward layers, residual connections and layer normalisation; masked-language pretraining learns sequence features.SourcesRetentive Network promotes efficient RNA language modeling of long sequences · Methods/Bidirectional encoder representation from Retentive Network (paragraph 1); Methods/Bidirectional encoder representation from Retentive Network (paragraph 2) |
| Biological inputs | RNA nucleotide sequencesSourcesRetentive Network promotes efficient RNA language modeling of long sequences · Introduction (paragraph 2); Methods/RNA tokenization (paragraph 1) |
| Outputs | RNA representations and downstream interaction, structure or classification predictionsSourcesRetentive Network promotes efficient RNA language modeling of long sequences · Methods/Fine-tuning strategies for downstream tasks (paragraph 1); Methods/RNA-RNA interaction prediction (paragraph 1) |
| Parameters | Approximately 12 million trainable parameters.SourcesRetentive Network promotes efficient RNA language modeling of long sequences · Methods/Fine-tuning strategies for downstream tasks (paragraph 2); Methods/RNAret model architecture (paragraph 1) |
| Known versions / configuration | 5-merSourcesRetentive Network promotes efficient RNA language modeling of long sequences · Table Tab3 (paragraph 1); Table Tab2 (paragraph 1) |
| Training data / fitting | 29.8 million RNA sequencesSourcesRetentive Network promotes efficient RNA language modeling of long sequences · Methods/Pretraining strategies (paragraph 1); Methods/Pretraining strategies (paragraph 3) |
| Context limits | A maximum input/context length for this exact evaluated configuration is not established by the inspected sources. · Not reported in inspected sourcesSources (2)Retentive Network promotes efficient RNA language modeling of long sequences; DrBlackZJU/RNAret README.md · Results/Self-supervised pretraining and feature extraction for RNA sequences; Methods/Bidirectional encoder representation from Retentive Network; Methods/RNAret model architecture; Methods/RNA tokenization; Methods/Pretraining strategies; Methods/Fine-tuning strategies for downstream tasks; Methods/RNA-RNA interaction prediction; Methods/RNA secondary structure prediction; inspected for explicit maximum input length (dataset lengths and family-wide limits are not substituted); README.md at pinned repository revision |
| Access | Official study implementation and usage documentation: https://github.com/DrBlackZJU/RNAret/blob/40ddab25fc038ba2b96bc9b9f88216abe38b2f64/README.md. This pinned documentation revision is not automatically the evaluated weight revision.SourcesDrBlackZJU/RNAret README.md · README.md; installation, model download and usage instructions |
| Code licence | CC0 1.0 (study repository code at the cited revision; this does not establish every dependency or historical checkpoint licence).SourcesDrBlackZJU/RNAret LICENSE · LICENSE; complete licence text |
| Weights licence | The inspected model-access documentation does not explicitly identify terms for this exact evaluated checkpoint or fitted head; repository code terms are shown separately. · Not reported in inspected sourcesSourcesDrBlackZJU/RNAret README.md · README.md; checkpoint/access documentation and licence scope |
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.
20 evidence rows matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|---|---|
| Diagram caption Conceptual input–method–output guide. Check the procedure text and linked evaluation for fitted components, additional inputs and exact settings. Individual claims | Retentive Network promotes efficient RNA language modeling of long sequences Methods/Bidirectional encoder representation from Retentive Network (paragraph 1); Methods/Bidirectional encoder representation from Retentive Network (paragraph 2) Version: journal full text in PMC | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Diagram steps ["RNA nucleotide sequences","RNAret","RNA representations and downstream interaction, structure or classification predictions"] Individual claims | Retentive Network promotes efficient RNA language modeling of long sequences Methods/Bidirectional encoder representation from Retentive Network (paragraph 1); Methods/Bidirectional encoder representation from Retentive Network (paragraph 2) Version: journal full text in PMC | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Diagram title Evaluated procedure (conceptual) Individual claims | Retentive Network promotes efficient RNA language modeling of long sequences Methods/Bidirectional encoder representation from Retentive Network (paragraph 1); Methods/Bidirectional encoder representation from Retentive Network (paragraph 2) Version: journal full text in PMC | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Model type Study-specific predictive method; this record is the paper-specific evaluated configuration. Individual claims | Retentive Network promotes efficient RNA language modeling of long sequences Methods/Bidirectional encoder representation from Retentive Network (paragraph 1); Methods/Bidirectional encoder representation from Retentive Network (paragraph 2) Version: journal full text in PMC | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Architecture / procedure Bidirectional retention blocks use multiscale retention, feed-forward layers, residual connections and layer normalisation; masked-language pretraining learns sequence features. Individual claims | Retentive Network promotes efficient RNA language modeling of long sequences Methods/Bidirectional encoder representation from Retentive Network (paragraph 1); Methods/Bidirectional encoder representation from Retentive Network (paragraph 2) Version: journal full text in PMC | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Weights licence The inspected model-access documentation does not explicitly identify terms for this exact evaluated checkpoint or fitted head; repository code terms are shown separately. Individual claims | DrBlackZJU/RNAret README.md README.md; checkpoint/access documentation and licence scope Version: 40ddab25fc038ba2b96bc9b9f88216abe38b2f64 | unreported automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Biological inputs RNA nucleotide sequences Individual claims | Retentive Network promotes efficient RNA language modeling of long sequences Introduction (paragraph 2); Methods/RNA tokenization (paragraph 1) Version: journal full text in PMC | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Outputs RNA representations and downstream interaction, structure or classification predictions Individual claims | Retentive Network promotes efficient RNA language modeling of long sequences Methods/Fine-tuning strategies for downstream tasks (paragraph 1); Methods/RNA-RNA interaction prediction (paragraph 1) Version: journal full text in PMC | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Parameters Approximately 12 million trainable parameters. Individual claims | Retentive Network promotes efficient RNA language modeling of long sequences Methods/Fine-tuning strategies for downstream tasks (paragraph 2); Methods/RNAret model architecture (paragraph 1) Version: journal full text in PMC | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Known versions / configuration 5-mer Individual claims | Retentive Network promotes efficient RNA language modeling of long sequences Table Tab3 (paragraph 1); Table Tab2 (paragraph 1) Version: journal full text in PMC | source checked automated source review · 2026-09-16 Audit detailsPrimary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
Sources and history
Release 2026-09-17-d277315f7d76 · Record review: needs review
3 source records and release history
- Retentive Network promotes efficient RNA language modeling of long sequences · Original source · journal full text in PMC
- DrBlackZJU/RNAret README.md · Original source · 40ddab25fc038ba2b96bc9b9f88216abe38b2f64
- DrBlackZJU/RNAret LICENSE · Original source · 40ddab25fc038ba2b96bc9b9f88216abe38b2f64
Technical metadata and extraction receipts
Stable ID: reported-model-7234658bc9c828
- areas
- rna-transcriptomes
- entity level
- method
- version
- 5-mer
- reported name
- RNAret
- historical missing metadata
- checkpoint revision: not_reported_in_legacy_extract; training data: not_reported_in_legacy_extract; licence: not_reported_in_legacy_extract
- metadata review scope
- historical_missing_metadata preserves the original discovery state. Current descriptive evidence and missingness are recorded in profile.facts; numerical-result review is separate.
- legacy kinds
- model
- entity classification
- review date: 2026-09-17; rationale: This source-scoped entry preserves the method/configuration actually named in an evaluation. It is neither a global family identity nor proof of an immutable checkpoint; the linked evaluation retains adaptation, fitting and scoring details.; source ids: rnaret-2026; source locator: Methods/Bidirectional encoder representation from Retentive Network (paragraph 1); Methods/Bidirectional encoder representation from Retentive Network (paragraph 2) | Methods/Bidirectional encoder representation from Retentive Network (paragraph 1); Methods/Bidirectional encoder representation from Retentive Network (paragraph 7); ambiguities: Configuration means the source-labelled evaluated identity. It does not establish missing checkpoint hashes, default settings or equivalence to same-named records in other papers.