Model type
Study-specific predictive method; this record is the paper-specific evaluated configuration.
Configuration
ADAR-GPT continual
ADAR-GPT is a supervised RNA-editing classifier obtained by continually fine-tuning GPT-4o-mini on marked RNA sequence windows.
Sources
ADAR-GPT: A continually fine-tuned language model for predicting A-to-I RNA editing sites · Results and Analysis (paragraph 1); Materials and Methods/Fine-Tuning Protocol. (paragraph 4)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
ADAR-GPT: A continually fine-tuned language model for predicting A-to-I RNA editing sites · Methodology/Model Training Approaches—Static vs. Continual Fine-Tuning. (paragraph 4); Contribution. (paragraph 1)At a glance
Biological inputs
201-nt RNA windows with the target adenosine explicitly marked
Sources
ADAR-GPT: A continually fine-tuned language model for predicting A-to-I RNA editing sites · Materials and Methods/Input Representation. (paragraph 1); Methodology/Data Collection and Preprocessing—Liver GTEx Dataset. (paragraph 5)Outputs
A-to-I editing-site classification at the study’s editing threshold
Sources
ADAR-GPT: A continually fine-tuned language model for predicting A-to-I RNA editing sites · Materials and Methods/Dataset Design and Labeling. (paragraph 1); Contribution. (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 |
|---|---|---|
| ADAR-GPT continual: A-to-I RNA editing site prediction Configuration: ADAR-GPT continualTask: A-to-I RNA editing site predictionDataset: liver editing sites Curriculum plus 15% fine-tuning; 201-nt sequence windows; decision threshold 0.5 Author-reported evaluation · Evaluation metadata: needs review | ||
| 0.763 F1 Unit: fraction · Direction: unknown | Uncertainty: not reported in legacy extract Scored: Not reported · Eligible: Not reported | source checkedADAR-GPT: A continually fine-tuned language model for predicting A-to-I RNA editing sites · Table 2, Adar-GPT (continual) row, F1 column Source checking is not independent reproduction. |
How it works
How the evaluated method works
The continual configuration first trains on progressively ordered editing-threshold data and then refines on sites meeting the 15% editing threshold. This is a task-specific fine-tuned pipeline, not an evaluation of the unadapted general-purpose model.
Sources
ADAR-GPT: A continually fine-tuned language model for predicting A-to-I RNA editing sites · Methodology/Model Training Approaches—Static vs. Continual Fine-Tuning. (paragraph 4); Contribution. (paragraph 1)What was evaluated
The linked evaluation record identifies ADAR-GPT continual: A-to-I RNA editing site prediction. Its dataset, split, adaptation and evidence origin remain attached to the reported results.
Sources
ADAR-GPT: A continually fine-tuned language model for predicting A-to-I RNA editing sites · The named evaluation’s methods and comparison table; exact preserved evaluation IDs: evaluation-lit-b4-005Strengths and limitations
Strengths and considerations
- Explicitly controls the target position and uses a threshold curriculum to adapt the classifier to the final editing definition.
Sources
ADAR-GPT: A continually fine-tuned language model for predicting A-to-I RNA editing sites · Previous Work (paragraph 5); Abstract (paragraph 2)
Limitations and conditions
- Evaluation is restricted to the study’s GTEx liver setting; broader tissue transfer and independent experimental validation remain separate questions.
Sources
ADAR-GPT: A continually fine-tuned language model for predicting A-to-I RNA editing sites · Abstract (paragraph 1); Discussion and Conclusions (paragraph 5)
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-1d2aa9880a1c77Specifications
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.SourcesADAR-GPT: A continually fine-tuned language model for predicting A-to-I RNA editing sites · Methodology/Model Training Approaches—Static vs. Continual Fine-Tuning. (paragraph 4); Contribution. (paragraph 1) |
| Architecture / procedure | The continual configuration first trains on progressively ordered editing-threshold data and then refines on sites meeting the 15% editing threshold. This is a task-specific fine-tuned pipeline, not an evaluation of the unadapted general-purpose model.SourcesADAR-GPT: A continually fine-tuned language model for predicting A-to-I RNA editing sites · Methodology/Model Training Approaches—Static vs. Continual Fine-Tuning. (paragraph 4); Contribution. (paragraph 1) |
| Biological inputs | 201-nt RNA windows with the target adenosine explicitly markedSourcesADAR-GPT: A continually fine-tuned language model for predicting A-to-I RNA editing sites · Materials and Methods/Input Representation. (paragraph 1); Methodology/Data Collection and Preprocessing—Liver GTEx Dataset. (paragraph 5) |
| Outputs | A-to-I editing-site classification at the study’s editing thresholdSourcesADAR-GPT: A continually fine-tuned language model for predicting A-to-I RNA editing sites · Materials and Methods/Dataset Design and Labeling. (paragraph 1); Contribution. (paragraph 1) |
| Parameters | The paper identifies GPT-4o-mini but does not disclose the backbone parameter count. · Not reported in inspected sourcesSourcesADAR-GPT: A continually fine-tuned language model for predicting A-to-I RNA editing sites · Results and Analysis (paragraph 1); Materials and Methods/Fine-Tuning Protocol. (paragraph 4) |
| Known versions / configuration | ADAR-GPT continual is the comparison-table label; that label does not specify an immutable weight revision. · Not reported in inspected sourcesSourcesADAR-GPT: A continually fine-tuned language model for predicting A-to-I RNA editing sites · Model identification in the comparison table and corresponding Methods; immutable checkpoint revision is not supplied by the table label. |
| Training data / fitting | GTEx liver data from 131 samples; curriculum thresholds 1%, 5%, 10% and 15%, followed by refinement on 15% sites.SourcesADAR-GPT: A continually fine-tuned language model for predicting A-to-I RNA editing sites · Methodology/Model Training Approaches—Static vs. Continual Fine-Tuning. (paragraph 4); Methodology/Data Collection and Preprocessing—Liver GTEx Dataset. (paragraph 3) |
| Context limits | 201 nucleotides: 100 upstream, central adenosine and 100 downstreamSourcesADAR-GPT: A continually fine-tuned language model for predicting A-to-I RNA editing sites · Methodology/Data Collection and Preprocessing—Liver GTEx Dataset. (paragraph 5); Table t03 (paragraph 1) |
| Access | Official study implementation and usage documentation: https://github.com/Scientific-Computing-Lab/ADAR-GPT/blob/c0fd23679922d91a45520455d4ca0202a5ca609f/README.md. This pinned documentation revision is not automatically the evaluated weight revision.SourcesScientific-Computing-Lab/ADAR-GPT README.md · README.md; installation, model download and usage instructions |
| Code licence | MIT (study repository code at the cited revision; this does not establish every dependency or historical checkpoint licence).SourcesScientific-Computing-Lab/ADAR-GPT 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 sourcesSourcesScientific-Computing-Lab/ADAR-GPT 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.
19 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 | ADAR-GPT: A continually fine-tuned language model for predicting A-to-I RNA editing sites Methodology/Model Training Approaches—Static vs. Continual Fine-Tuning. (paragraph 4); Contribution. (paragraph 1) Version: version of record | 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 ["201-nt RNA windows with the target adenosine explicitly marked","ADAR-GPT continual","A-to-I editing-site classification at the study’s editing threshold"] Individual claims | ADAR-GPT: A continually fine-tuned language model for predicting A-to-I RNA editing sites Methodology/Model Training Approaches—Static vs. Continual Fine-Tuning. (paragraph 4); Contribution. (paragraph 1) Version: version of record | 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 | ADAR-GPT: A continually fine-tuned language model for predicting A-to-I RNA editing sites Methodology/Model Training Approaches—Static vs. Continual Fine-Tuning. (paragraph 4); Contribution. (paragraph 1) Version: version of record | 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 | ADAR-GPT: A continually fine-tuned language model for predicting A-to-I RNA editing sites Methodology/Model Training Approaches—Static vs. Continual Fine-Tuning. (paragraph 4); Contribution. (paragraph 1) Version: version of record | 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 The continual configuration first trains on progressively ordered editing-threshold data and then refines on sites meeting the 15% editing threshold. This is a task-specific fine-tuned pipeline, not an evaluation of the unadapted general-purpose model. Individual claims | ADAR-GPT: A continually fine-tuned language model for predicting A-to-I RNA editing sites Methodology/Model Training Approaches—Static vs. Continual Fine-Tuning. (paragraph 4); Contribution. (paragraph 1) Version: version of record | 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 | Scientific-Computing-Lab/ADAR-GPT README.md README.md; checkpoint/access documentation and licence scope Version: c0fd23679922d91a45520455d4ca0202a5ca609f | 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 201-nt RNA windows with the target adenosine explicitly marked Individual claims | ADAR-GPT: A continually fine-tuned language model for predicting A-to-I RNA editing sites Materials and Methods/Input Representation. (paragraph 1); Methodology/Data Collection and Preprocessing—Liver GTEx Dataset. (paragraph 5) Version: version of record | 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 A-to-I editing-site classification at the study’s editing threshold Individual claims | ADAR-GPT: A continually fine-tuned language model for predicting A-to-I RNA editing sites Materials and Methods/Dataset Design and Labeling. (paragraph 1); Contribution. (paragraph 1) Version: version of record | 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 The paper identifies GPT-4o-mini but does not disclose the backbone parameter count. Individual claims | ADAR-GPT: A continually fine-tuned language model for predicting A-to-I RNA editing sites Results and Analysis (paragraph 1); Materials and Methods/Fine-Tuning Protocol. (paragraph 4) Version: version of record | 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 |
| Known versions / configuration ADAR-GPT continual is the comparison-table label; that label does not specify an immutable weight revision. Individual claims | ADAR-GPT: A continually fine-tuned language model for predicting A-to-I RNA editing sites Model identification in the comparison table and corresponding Methods; immutable checkpoint revision is not supplied by the table label. Version: version of record | 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 |
Sources and history
Release 2026-09-17-d277315f7d76 · Record review: needs review
3 source records and release history
- ADAR-GPT: A continually fine-tuned language model for predicting A-to-I RNA editing sites · Original source · version of record
- Scientific-Computing-Lab/ADAR-GPT README.md · Original source · c0fd23679922d91a45520455d4ca0202a5ca609f
- Scientific-Computing-Lab/ADAR-GPT LICENSE · Original source · c0fd23679922d91a45520455d4ca0202a5ca609f
Technical metadata and extraction receipts
Stable ID: reported-model-1d2aa9880a1c77
- areas
- rna-transcriptomes
- entity level
- method
- version
- Not reported
- reported name
- ADAR-GPT continual
- historical missing metadata
- version: not_reported_in_legacy_extract; 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: adar-gpt-editing-2026; source locator: Methodology/Model Training Approaches—Static vs. Continual Fine-Tuning. (paragraph 4); Contribution. (paragraph 1) | Results and Analysis (paragraph 1); Materials and Methods/Fine-Tuning Protocol. (paragraph 4); 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.