Model type
DNA state-space model; this record is the paper-specific evaluated configuration.
Configuration
Caduceus
This genomic model is adapted to predict G-quadruplex-associated sequence regions.
Sources
Benchmarking DNA large language models on quadruplexes · Discussion and conclusions (paragraph 6); Results/LLM performance at the genome-wide level (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
Benchmarking DNA large language models on quadruplexes · Materials and methods/Data preparation (paragraph 2); Results/Interpretation of LLMs (paragraph 2)At a glance
Biological inputs
DNA sequences for G-quadruplex classification and genomic scanning
Sources
Benchmarking DNA large language models on quadruplexes · Results/LLM comparison on quadruplex datasets (paragraph 1); Materials and methods/Data preparation (paragraph 1)Outputs
G4-associated sequence predictions
Sources
Benchmarking DNA large language models on quadruplexes · Introduction (paragraph 3); CRediT authorship contribution statement (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 |
|---|---|---|
| Caduceus: G-quadruplex classification Pretrained model evaluated on KEx as reported in Table 5. Independent external evaluation · Evaluation metadata: needs review | ||
| 95.0 Accuracy Unit: % · Direction: unknown | Uncertainty: ± 0.5 Scored: Not reported · Eligible: Not reported | source checkedBenchmarking DNA large language models on quadruplexes · Table 5, Caduceus (8 M) row, Accuracy column Source checking is not independent reproduction. |
How it works
How the evaluated method works
The study fine-tunes genomic sequence models for G4 classification and applies them to whole-genome annotation, retaining each model’s native tokenisation.
Sources
Benchmarking DNA large language models on quadruplexes · Materials and methods/Data preparation (paragraph 2); Results/Interpretation of LLMs (paragraph 2)Underlying method and version boundaries
Caduceus exposes distinct Ph and PS configurations. The documented Ph-131k checkpoint uses 16 layers, width 256 and reverse-complement data augmentation; PS implements reverse-complement equivariance. These training choices are not interchangeable.
Sources
kuleshov-group/caduceus README.md · README.md; introduction, model description, pretrained-model and usage sections at pinned revisionWhat was evaluated
The linked evaluation record identifies Caduceus: G-quadruplex classification. Its dataset, split, adaptation and evidence origin remain attached to the reported results.
Sources
Benchmarking DNA large language models on quadruplexes · The named evaluation’s methods and comparison table; exact preserved evaluation IDs: evaluation-lit-006Strengths and limitations
Strengths and considerations
- Compares transformer, long-convolution and state-space architectures on a specialised non-B-DNA task.
Sources
Benchmarking DNA large language models on quadruplexes · Introduction (paragraph 1); Abstract (paragraph 1)
Limitations and conditions
- Predicted G4 sequence potential is not a direct measurement of structure formation in a particular cellular context.
Sources
Benchmarking DNA large language models on quadruplexes · Introduction (paragraph 6); Discussion and conclusions (paragraph 1)
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-abc19288fe9009Specifications
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 | DNA state-space model; this record is the paper-specific evaluated configuration.Sourceskuleshov-group/caduceus README.md · README.md model description |
| Architecture / procedure | The study fine-tunes genomic sequence models for G4 classification and applies them to whole-genome annotation, retaining each model’s native tokenisation.SourcesBenchmarking DNA large language models on quadruplexes · Materials and methods/Data preparation (paragraph 2); Results/Interpretation of LLMs (paragraph 2) |
| Biological inputs | DNA sequences for G-quadruplex classification and genomic scanningSourcesBenchmarking DNA large language models on quadruplexes · Results/LLM comparison on quadruplex datasets (paragraph 1); Materials and methods/Data preparation (paragraph 1) |
| Outputs | G4-associated sequence predictionsSourcesBenchmarking DNA large language models on quadruplexes · Introduction (paragraph 3); CRediT authorship contribution statement (paragraph 1) |
| Parameters | 8 million parameters, as identified for this rowSourcesBenchmarking DNA large language models on quadruplexes · Materials and methods/Low Rank Adaptation (LoRA) (paragraph 1); Discussion and conclusions (paragraph 3) |
| Known versions / configuration | 8MSourcesBenchmarking DNA large language models on quadruplexes · Table tbl0030 (paragraph 1); Table tbl0025 (paragraph 1) |
| Training data / fitting | Task-specific fine-tuning for approximately four to ten epochs, depending on model performance, with gradient accumulation.SourcesBenchmarking DNA large language models on quadruplexes · Results/LLM comparison on quadruplex datasets (paragraph 5); Materials and methods/Fine-tuning (paragraph 1) |
| 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)Benchmarking DNA large language models on quadruplexes; kuleshov-group/caduceus README.md · Materials and methods/Data preparation; Materials and methods/Tokenization for G4s; Materials and methods/Metrics of evaluation; Materials and methods/Fine-tuning; Materials and methods/Low Rank Adaptation (LoRA); inspected for explicit maximum input length (dataset lengths and family-wide limits are not substituted); README.md at pinned repository revision |
| Access | Official upstream implementation and usage documentation: https://github.com/kuleshov-group/caduceus/blob/0060a6d8079b6a040fc55d505e15972a327b70a6/README.md. This pinned documentation revision is not automatically the evaluated weight revision.Sourceskuleshov-group/caduceus README.md · README.md; installation, model download and usage instructions |
| Code licence | Apache 2.0 (upstream repository code at the cited revision; this does not establish every dependency or historical checkpoint licence).Sourceskuleshov-group/caduceus 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 sourcesSourceskuleshov-group/caduceus 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.
21 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 | Benchmarking DNA large language models on quadruplexes Materials and methods/Data preparation (paragraph 2); Results/Interpretation of LLMs (paragraph 2) 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 ["DNA sequences for G-quadruplex classification and genomic scanning","Caduceus","G4-associated sequence predictions"] Individual claims | Benchmarking DNA large language models on quadruplexes Materials and methods/Data preparation (paragraph 2); Results/Interpretation of LLMs (paragraph 2) 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 | Benchmarking DNA large language models on quadruplexes Materials and methods/Data preparation (paragraph 2); Results/Interpretation of LLMs (paragraph 2) 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 DNA state-space model; this record is the paper-specific evaluated configuration. Individual claims | kuleshov-group/caduceus README.md README.md model description Version: 0060a6d8079b6a040fc55d505e15972a327b70a6 | 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 study fine-tunes genomic sequence models for G4 classification and applies them to whole-genome annotation, retaining each model’s native tokenisation. Individual claims | Benchmarking DNA large language models on quadruplexes Materials and methods/Data preparation (paragraph 2); Results/Interpretation of LLMs (paragraph 2) 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 | kuleshov-group/caduceus README.md README.md; checkpoint/access documentation and licence scope Version: 0060a6d8079b6a040fc55d505e15972a327b70a6 | 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 DNA sequences for G-quadruplex classification and genomic scanning Individual claims | Benchmarking DNA large language models on quadruplexes Results/LLM comparison on quadruplex datasets (paragraph 1); Materials and methods/Data preparation (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 |
| Outputs G4-associated sequence predictions Individual claims | Benchmarking DNA large language models on quadruplexes Introduction (paragraph 3); CRediT authorship contribution statement (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 8 million parameters, as identified for this row Individual claims | Benchmarking DNA large language models on quadruplexes Materials and methods/Low Rank Adaptation (LoRA) (paragraph 1); Discussion and conclusions (paragraph 3) 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 |
| Known versions / configuration 8M Individual claims | Benchmarking DNA large language models on quadruplexes Table tbl0030 (paragraph 1); Table tbl0025 (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 |
Sources and history
Release 2026-09-17-d277315f7d76 · Record review: needs review
3 source records and release history
- Benchmarking DNA large language models on quadruplexes · Original source · version of record
- kuleshov-group/caduceus README.md · Original source · 0060a6d8079b6a040fc55d505e15972a327b70a6
- kuleshov-group/caduceus LICENSE · Original source · 0060a6d8079b6a040fc55d505e15972a327b70a6
Technical metadata and extraction receipts
Stable ID: reported-model-abc19288fe9009
- areas
- dna-genomes
- entity level
- method
- version
- 8M
- reported name
- Caduceus
- 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: quadruplex-llm-benchmark-2025; evidence-reported-base-caduceus-readme-md; source locator: Materials and methods/Data preparation (paragraph 2); Results/Interpretation of LLMs (paragraph 2) | README.md model description | Discussion and conclusions (paragraph 6); Results/LLM performance at the genome-wide level (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.