0.876 accuracy
DNABERT-2 · accuracy · DART-Eval cCREs versus matched shuffled controls
- Tested model
- DNABERT-2
- Task or benchmark
- regulatory element identification
- Dataset
- DART-Eval cCREs versus matched shuffled controls
- Procedure
- zero-shot likelihood ranking: higher likelihood for cCRE than matched control
- Evaluation
- DNABERT-2: regulatory element identification
- Evidence
- Independent external evaluation · source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 3 (PDF page 5), DNABERT-2 row, Zero-Shot Accuracy column
A source-checked result verifies the numerical transcription, not every model or protocol detail. Evaluation metadata: needs review. Source checked does not mean independently reproduced.
Evaluation results
Release 2026-09-16-d74d282221a9 · 1 evaluation · 1 metric row. Different protocols are not a single leaderboard.
| Metric and finding | Coverage and uncertainty | Evidence |
|---|---|---|
| DNABERT-2: regulatory element identification Model: DNABERT-2 · Benchmark: regulatory element identification · Dataset: DART-Eval cCREs versus matched shuffled controls zero-shot likelihood ranking: higher likelihood for cCRE than matched control Independent external evaluation · Evaluation metadata: needs review | ||
| 0.876 accuracy Unit: fraction · Direction: unknown | Uncertainty: not reported in legacy extract Scored: Not reported · Eligible: Not reported | source checkedDART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Table 3 (PDF page 5), DNABERT-2 row, Zero-Shot Accuracy column Source checking is not independent reproduction. |
Sources and history
Release 2026-09-16-d74d282221a9 · Record review: source checked
- DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA · Original source · NeurIPS 2024 Datasets and Benchmarks Track proceedings
Technical metadata and extraction receipts
Stable ID: b2-dart-eval-regulatory-2024
- areas
- dna-genomes
- tasks
- regulatory element identification
- printed value
- 0.876
- numeric value
- 0.876
- metric
- accuracy
- metric direction
- unknown
- unit
- fraction
- uncertainty
- Not reported
- source locator
- Table 3 (PDF page 5), DNABERT-2 row, Zero-Shot Accuracy column
- review
- method: independent_ai_table_review; reviewer: Codex secondary table review; reviewed at: 2026-09-16T10:38:57.558203+00:00; notes: Inspected the pinned NeurIPS primary PDF table and explanatory text. The DNABERT-2 row reports 0.876 under Zero-Shot Accuracy. The caption defines this as pairwise prioritization of positives over matched controls, distinct from supervised absolute accuracy.; evidence: DNABERT-2; zero-shot accuracy 0.876; probed absolute/paired 0.847/0.943; fine-tuned absolute/paired 0.913/0.973.; artifact sha256: e5aee5b1f7cc6fd961b1d2a131d02cf243b79e091d5e418fbabee7fde9b39b22; retrieval url: https://proceedings.neurips.cc/paper_files/paper/2024/file/71998bfc3217ffe1cca1ee084dfadadd-Paper-Datasets_and_Benchmarks_Track.pdf
- legacy id
- b2-dart-eval-regulatory-2024
- legacy row
- id: b2-dart-eval-regulatory-2024; paper id: dart-eval-regulatory-2024; domain id: dna-genomes; task: regulatory element identification; model: DNABERT-2; model version: not stated in table; dataset: DART-Eval cCREs versus matched shuffled controls; dataset version: Not reported; split: paper evaluation; metric: accuracy; value: 0.876; unit: fraction; uncertainty: Not reported; protocol: zero-shot likelihood ranking: higher likelihood for cCRE than matched control; source locator: Table 3, DNABERT-2 row, Zero-Shot Accuracy column; source url: https://proceedings.neurips.cc/paper_files/paper/2024/file/71998bfc3217ffe1cca1ee084dfadadd-Paper-Datasets_and_Benchmarks_Track.pdf; evaluation origin: independent_paper; reviewed utc: 2026-09-15T23:33:26Z
- missing metadata
- dataset version: not_reported_in_legacy_extract; uncertainty: not_reported_in_legacy_extract
Related records
- evaluation: DNABERT-2: regulatory element identification
- subject: Reported accuracy for DNABERT-2