Protocol

Wang et al. 2026 zero-shot secondary structure, TS test set (70 PDB structures), attention, F1 score (mean over structures)

Zero-shot: symmetrised, APC-corrected attention map from one head-layer and threshold chosen on VL1 by F1; F1 score reported as mean over structures (Supplementary Table 4).

32 evaluations · 32 results

Overview

An explanatory profile has not yet been reviewed for this record.

Consult the linked sources for architecture or protocol details. Missing evidence is not evidence of a missing capability.

32 recorded evaluations, 32 metric rows. A comparison chart has not yet been validated for these results. The table retains the individual findings and their sources.

View coverage and remaining gaps across all benchmarks

Results

Results are available, but no reviewed comparison panel is linked in this release.

All evaluations

32 evaluations · 32 results. Different protocols are not a single leaderboard.

Protocol: Wang et al. 2026 zero-shot secondary structure, TS test set (70 PDB structures), attention, F1 score (mean over structures) · Dataset: TS test set (70 PDB structures) (Wang et al. 2026)

Sorted by F1 (attention; mean over structures) (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	F1 (attention; mean over structures)
RNA-MSM (zero-shot, Wang et al. 2026)Evaluation	0.582±0.217 (best in column)
AIDO.RNA (1.6B) (zero-shot, Wang et al. 2026)Evaluation	0.508±0.263
MP-RNA (zero-shot, Wang et al. 2026)Evaluation	0.507±0.201
RNA-km (zero-shot, Wang et al. 2026)Evaluation	0.492±0.236
RiNALMo (giga) (zero-shot, Wang et al. 2026)Evaluation	0.414±0.255
AIDO.RNA (650M) (zero-shot, Wang et al. 2026)Evaluation	0.402±0.234
RiNALMo (mega) (zero-shot, Wang et al. 2026)Evaluation	0.342±0.225
ProtRNA (zero-shot, Wang et al. 2026)Evaluation	0.333±0.236
Uni-RNA (L16) (zero-shot, Wang et al. 2026)Evaluation	0.307±0.197
ERNIE-RNA (zero-shot, Wang et al. 2026)Evaluation	0.300±0.234
RNAErnie (zero-shot, Wang et al. 2026)Evaluation	0.288±0.223
RiNALMo (micro) (zero-shot, Wang et al. 2026)Evaluation	0.268±0.184
RNA-FM (zero-shot, Wang et al. 2026)Evaluation	0.267±0.239
Uni-RNA (L12) (zero-shot, Wang et al. 2026)Evaluation	0.258±0.198
Uni-RNA (L8) (zero-shot, Wang et al. 2026)Evaluation	0.257±0.197
LucaOne (G36.8M) (zero-shot, Wang et al. 2026)Evaluation	0.253±0.232
LucaOne (D17.6M) (zero-shot, Wang et al. 2026)Evaluation	0.224±0.194
LucaOne (D36.0M) (zero-shot, Wang et al. 2026)Evaluation	0.211±0.200
PlantRNA-FM (zero-shot, Wang et al. 2026)Evaluation	0.160±0.137
RFamLlama (Large) (zero-shot, Wang et al. 2026)Evaluation	0.052±0.057
SpliceBERT (zero-shot, Wang et al. 2026)Evaluation	0.043±0.039
RFamLlama (Base) (zero-shot, Wang et al. 2026)Evaluation	0.042±0.040
LAMAR (2k) (zero-shot, Wang et al. 2026)Evaluation	0.037±0.021
LAMAR (4k) (zero-shot, Wang et al. 2026)Evaluation	0.033±0.040
UTR_LM (zero-shot, Wang et al. 2026)Evaluation	0.029±0.032
3UTRBERT (K5) (zero-shot, Wang et al. 2026)Evaluation	0.023±0.034
3UTRBERT (K3) (zero-shot, Wang et al. 2026)Evaluation	0.022±0.027
3UTRBERT (K6) (zero-shot, Wang et al. 2026)Evaluation	0.022±0.012
RNABERT (zero-shot, Wang et al. 2026)Evaluation	0.022±0.032
3UTRBERT (K4) (zero-shot, Wang et al. 2026)Evaluation	0.018±0.027
DNABERT (k=6) (zero-shot, Wang et al. 2026)Evaluation	0.017±0.021
DNABERT (k=3) (zero-shot, Wang et al. 2026)Evaluation	0.011±0.019

All 32 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: 3UTRBERT (K3) (zero-shot, Wang et al. 2026)	Protocol: Wang et al. 2026 zero-shot secondary structure, TS test set (70 PDB structures), attention, F1 score (mean over structures) Dataset: TS test set (70 PDB structures) (Wang et al. 2026)	0.022±0.027 F1 (attention; mean over structures) unitless · higher Uncertainty: SD 0.027 Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source 3UTRBERT (K3) zero-shot secondary structure on TS test set (70 PDB structures) (attention, F1 score, mean) lit-rna-20261011-rnalm-zeroshot-protocol-t4-f1-score-mean Aggregation: Not reported Zero-shot benchmarking of RNA language models: Supplementary Information; Zero-shot benchmarking of RNA language models in structural, functional, and evolutionary learning · Supplementary Table 4, row '3UTRBERT (K3)', column 'F1 score' (Mean ± Std)
Configuration: 3UTRBERT (K4) (zero-shot, Wang et al. 2026)	Protocol: Wang et al. 2026 zero-shot secondary structure, TS test set (70 PDB structures), attention, F1 score (mean over structures) Dataset: TS test set (70 PDB structures) (Wang et al. 2026)	0.018±0.027 F1 (attention; mean over structures) unitless · higher Uncertainty: SD 0.027 Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source 3UTRBERT (K4) zero-shot secondary structure on TS test set (70 PDB structures) (attention, F1 score, mean) lit-rna-20261011-rnalm-zeroshot-protocol-t4-f1-score-mean Aggregation: Not reported Zero-shot benchmarking of RNA language models: Supplementary Information; Zero-shot benchmarking of RNA language models in structural, functional, and evolutionary learning · Supplementary Table 4, row '3UTRBERT (K4)', column 'F1 score' (Mean ± Std)
Configuration: 3UTRBERT (K5) (zero-shot, Wang et al. 2026)	Protocol: Wang et al. 2026 zero-shot secondary structure, TS test set (70 PDB structures), attention, F1 score (mean over structures) Dataset: TS test set (70 PDB structures) (Wang et al. 2026)	0.023±0.034 F1 (attention; mean over structures) unitless · higher Uncertainty: SD 0.034 Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source 3UTRBERT (K5) zero-shot secondary structure on TS test set (70 PDB structures) (attention, F1 score, mean) lit-rna-20261011-rnalm-zeroshot-protocol-t4-f1-score-mean Aggregation: Not reported Zero-shot benchmarking of RNA language models: Supplementary Information; Zero-shot benchmarking of RNA language models in structural, functional, and evolutionary learning · Supplementary Table 4, row '3UTRBERT (K5)', column 'F1 score' (Mean ± Std)
Configuration: 3UTRBERT (K6) (zero-shot, Wang et al. 2026)	Protocol: Wang et al. 2026 zero-shot secondary structure, TS test set (70 PDB structures), attention, F1 score (mean over structures) Dataset: TS test set (70 PDB structures) (Wang et al. 2026)	0.022±0.012 F1 (attention; mean over structures) unitless · higher Uncertainty: SD 0.012 Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source 3UTRBERT (K6) zero-shot secondary structure on TS test set (70 PDB structures) (attention, F1 score, mean) lit-rna-20261011-rnalm-zeroshot-protocol-t4-f1-score-mean Aggregation: Not reported Zero-shot benchmarking of RNA language models: Supplementary Information; Zero-shot benchmarking of RNA language models in structural, functional, and evolutionary learning · Supplementary Table 4, row '3UTRBERT (K6)', column 'F1 score' (Mean ± Std)
Configuration: AIDO.RNA (1.6B) (zero-shot, Wang et al. 2026)	Protocol: Wang et al. 2026 zero-shot secondary structure, TS test set (70 PDB structures), attention, F1 score (mean over structures) Dataset: TS test set (70 PDB structures) (Wang et al. 2026)	0.508±0.263 F1 (attention; mean over structures) unitless · higher Uncertainty: SD 0.263 Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source AIDO.RNA (1.6B) zero-shot secondary structure on TS test set (70 PDB structures) (attention, F1 score, mean) lit-rna-20261011-rnalm-zeroshot-protocol-t4-f1-score-mean Aggregation: Not reported Zero-shot benchmarking of RNA language models: Supplementary Information; Zero-shot benchmarking of RNA language models in structural, functional, and evolutionary learning · Supplementary Table 4, row 'AIDO.RNA (1.6B)', column 'F1 score' (Mean ± Std)
Configuration: AIDO.RNA (650M) (zero-shot, Wang et al. 2026)	Protocol: Wang et al. 2026 zero-shot secondary structure, TS test set (70 PDB structures), attention, F1 score (mean over structures) Dataset: TS test set (70 PDB structures) (Wang et al. 2026)	0.402±0.234 F1 (attention; mean over structures) unitless · higher Uncertainty: SD 0.234 Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source AIDO.RNA (650M) zero-shot secondary structure on TS test set (70 PDB structures) (attention, F1 score, mean) lit-rna-20261011-rnalm-zeroshot-protocol-t4-f1-score-mean Aggregation: Not reported Zero-shot benchmarking of RNA language models: Supplementary Information; Zero-shot benchmarking of RNA language models in structural, functional, and evolutionary learning · Supplementary Table 4, row 'AIDO.RNA (650M)', column 'F1 score' (Mean ± Std)
Configuration: DNABERT (k=3) (zero-shot, Wang et al. 2026)	Protocol: Wang et al. 2026 zero-shot secondary structure, TS test set (70 PDB structures), attention, F1 score (mean over structures) Dataset: TS test set (70 PDB structures) (Wang et al. 2026)	0.011±0.019 F1 (attention; mean over structures) unitless · higher Uncertainty: SD 0.019 Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source DNABERT (k=3) zero-shot secondary structure on TS test set (70 PDB structures) (attention, F1 score, mean) lit-rna-20261011-rnalm-zeroshot-protocol-t4-f1-score-mean Aggregation: Not reported Zero-shot benchmarking of RNA language models: Supplementary Information; Zero-shot benchmarking of RNA language models in structural, functional, and evolutionary learning · Supplementary Table 4, row 'DNABERT (k=3)', column 'F1 score' (Mean ± Std)
Configuration: DNABERT (k=6) (zero-shot, Wang et al. 2026)	Protocol: Wang et al. 2026 zero-shot secondary structure, TS test set (70 PDB structures), attention, F1 score (mean over structures) Dataset: TS test set (70 PDB structures) (Wang et al. 2026)	0.017±0.021 F1 (attention; mean over structures) unitless · higher Uncertainty: SD 0.021 Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source DNABERT (k=6) zero-shot secondary structure on TS test set (70 PDB structures) (attention, F1 score, mean) lit-rna-20261011-rnalm-zeroshot-protocol-t4-f1-score-mean Aggregation: Not reported Zero-shot benchmarking of RNA language models: Supplementary Information; Zero-shot benchmarking of RNA language models in structural, functional, and evolutionary learning · Supplementary Table 4, row 'DNABERT (k=6)', column 'F1 score' (Mean ± Std)
Configuration: ERNIE-RNA (zero-shot, Wang et al. 2026)	Protocol: Wang et al. 2026 zero-shot secondary structure, TS test set (70 PDB structures), attention, F1 score (mean over structures) Dataset: TS test set (70 PDB structures) (Wang et al. 2026)	0.300±0.234 F1 (attention; mean over structures) unitless · higher Uncertainty: SD 0.234 Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source ERNIE-RNA zero-shot secondary structure on TS test set (70 PDB structures) (attention, F1 score, mean) lit-rna-20261011-rnalm-zeroshot-protocol-t4-f1-score-mean Aggregation: Not reported Zero-shot benchmarking of RNA language models: Supplementary Information; Zero-shot benchmarking of RNA language models in structural, functional, and evolutionary learning · Supplementary Table 4, row 'ERNIE-RNA', column 'F1 score' (Mean ± Std)
Configuration: LAMAR (2k) (zero-shot, Wang et al. 2026)	Protocol: Wang et al. 2026 zero-shot secondary structure, TS test set (70 PDB structures), attention, F1 score (mean over structures) Dataset: TS test set (70 PDB structures) (Wang et al. 2026)	0.037±0.021 F1 (attention; mean over structures) unitless · higher Uncertainty: SD 0.021 Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source LAMAR (2k) zero-shot secondary structure on TS test set (70 PDB structures) (attention, F1 score, mean) lit-rna-20261011-rnalm-zeroshot-protocol-t4-f1-score-mean Aggregation: Not reported Zero-shot benchmarking of RNA language models: Supplementary Information; Zero-shot benchmarking of RNA language models in structural, functional, and evolutionary learning · Supplementary Table 4, row 'LAMAR (2k)', column 'F1 score' (Mean ± Std)
Configuration: LAMAR (4k) (zero-shot, Wang et al. 2026)	Protocol: Wang et al. 2026 zero-shot secondary structure, TS test set (70 PDB structures), attention, F1 score (mean over structures) Dataset: TS test set (70 PDB structures) (Wang et al. 2026)	0.033±0.040 F1 (attention; mean over structures) unitless · higher Uncertainty: SD 0.040 Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source LAMAR (4k) zero-shot secondary structure on TS test set (70 PDB structures) (attention, F1 score, mean) lit-rna-20261011-rnalm-zeroshot-protocol-t4-f1-score-mean Aggregation: Not reported Zero-shot benchmarking of RNA language models: Supplementary Information; Zero-shot benchmarking of RNA language models in structural, functional, and evolutionary learning · Supplementary Table 4, row 'LAMAR (4k)', column 'F1 score' (Mean ± Std)
Configuration: LucaOne (D17.6M) (zero-shot, Wang et al. 2026)	Protocol: Wang et al. 2026 zero-shot secondary structure, TS test set (70 PDB structures), attention, F1 score (mean over structures) Dataset: TS test set (70 PDB structures) (Wang et al. 2026)	0.224±0.194 F1 (attention; mean over structures) unitless · higher Uncertainty: SD 0.194 Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source LucaOne (D17.6M) zero-shot secondary structure on TS test set (70 PDB structures) (attention, F1 score, mean) lit-rna-20261011-rnalm-zeroshot-protocol-t4-f1-score-mean Aggregation: Not reported Zero-shot benchmarking of RNA language models: Supplementary Information; Zero-shot benchmarking of RNA language models in structural, functional, and evolutionary learning · Supplementary Table 4, row 'LucaOne (D17.6M)', column 'F1 score' (Mean ± Std)
Configuration: LucaOne (D36.0M) (zero-shot, Wang et al. 2026)	Protocol: Wang et al. 2026 zero-shot secondary structure, TS test set (70 PDB structures), attention, F1 score (mean over structures) Dataset: TS test set (70 PDB structures) (Wang et al. 2026)	0.211±0.200 F1 (attention; mean over structures) unitless · higher Uncertainty: SD 0.200 Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source LucaOne (D36.0M) zero-shot secondary structure on TS test set (70 PDB structures) (attention, F1 score, mean) lit-rna-20261011-rnalm-zeroshot-protocol-t4-f1-score-mean Aggregation: Not reported Zero-shot benchmarking of RNA language models: Supplementary Information; Zero-shot benchmarking of RNA language models in structural, functional, and evolutionary learning · Supplementary Table 4, row 'LucaOne (D36.0M)', column 'F1 score' (Mean ± Std)
Configuration: LucaOne (G36.8M) (zero-shot, Wang et al. 2026)	Protocol: Wang et al. 2026 zero-shot secondary structure, TS test set (70 PDB structures), attention, F1 score (mean over structures) Dataset: TS test set (70 PDB structures) (Wang et al. 2026)	0.253±0.232 F1 (attention; mean over structures) unitless · higher Uncertainty: SD 0.232 Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source LucaOne (G36.8M) zero-shot secondary structure on TS test set (70 PDB structures) (attention, F1 score, mean) lit-rna-20261011-rnalm-zeroshot-protocol-t4-f1-score-mean Aggregation: Not reported Zero-shot benchmarking of RNA language models: Supplementary Information; Zero-shot benchmarking of RNA language models in structural, functional, and evolutionary learning · Supplementary Table 4, row 'LucaOne (G36.8M)', column 'F1 score' (Mean ± Std)
Configuration: MP-RNA (zero-shot, Wang et al. 2026)	Protocol: Wang et al. 2026 zero-shot secondary structure, TS test set (70 PDB structures), attention, F1 score (mean over structures) Dataset: TS test set (70 PDB structures) (Wang et al. 2026)	0.507±0.201 F1 (attention; mean over structures) unitless · higher Uncertainty: SD 0.201 Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MP-RNA zero-shot secondary structure on TS test set (70 PDB structures) (attention, F1 score, mean) lit-rna-20261011-rnalm-zeroshot-protocol-t4-f1-score-mean Aggregation: Not reported Zero-shot benchmarking of RNA language models: Supplementary Information; Zero-shot benchmarking of RNA language models in structural, functional, and evolutionary learning · Supplementary Table 4, row 'MP-RNA', column 'F1 score' (Mean ± Std)
Configuration: PlantRNA-FM (zero-shot, Wang et al. 2026)	Protocol: Wang et al. 2026 zero-shot secondary structure, TS test set (70 PDB structures), attention, F1 score (mean over structures) Dataset: TS test set (70 PDB structures) (Wang et al. 2026)	0.160±0.137 F1 (attention; mean over structures) unitless · higher Uncertainty: SD 0.137 Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source PlantRNA-FM zero-shot secondary structure on TS test set (70 PDB structures) (attention, F1 score, mean) lit-rna-20261011-rnalm-zeroshot-protocol-t4-f1-score-mean Aggregation: Not reported Zero-shot benchmarking of RNA language models: Supplementary Information; Zero-shot benchmarking of RNA language models in structural, functional, and evolutionary learning · Supplementary Table 4, row 'PlantRNA-FM', column 'F1 score' (Mean ± Std)
Configuration: ProtRNA (zero-shot, Wang et al. 2026)	Protocol: Wang et al. 2026 zero-shot secondary structure, TS test set (70 PDB structures), attention, F1 score (mean over structures) Dataset: TS test set (70 PDB structures) (Wang et al. 2026)	0.333±0.236 F1 (attention; mean over structures) unitless · higher Uncertainty: SD 0.236 Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source ProtRNA zero-shot secondary structure on TS test set (70 PDB structures) (attention, F1 score, mean) lit-rna-20261011-rnalm-zeroshot-protocol-t4-f1-score-mean Aggregation: Not reported Zero-shot benchmarking of RNA language models: Supplementary Information; Zero-shot benchmarking of RNA language models in structural, functional, and evolutionary learning · Supplementary Table 4, row 'ProtRNA', column 'F1 score' (Mean ± Std)
Configuration: RFamLlama (Base) (zero-shot, Wang et al. 2026)	Protocol: Wang et al. 2026 zero-shot secondary structure, TS test set (70 PDB structures), attention, F1 score (mean over structures) Dataset: TS test set (70 PDB structures) (Wang et al. 2026)	0.042±0.040 F1 (attention; mean over structures) unitless · higher Uncertainty: SD 0.040 Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source RFamLlama (Base) zero-shot secondary structure on TS test set (70 PDB structures) (attention, F1 score, mean) lit-rna-20261011-rnalm-zeroshot-protocol-t4-f1-score-mean Aggregation: Not reported Zero-shot benchmarking of RNA language models: Supplementary Information; Zero-shot benchmarking of RNA language models in structural, functional, and evolutionary learning · Supplementary Table 4, row 'RFamLlama (Base)', column 'F1 score' (Mean ± Std)
Configuration: RFamLlama (Large) (zero-shot, Wang et al. 2026)	Protocol: Wang et al. 2026 zero-shot secondary structure, TS test set (70 PDB structures), attention, F1 score (mean over structures) Dataset: TS test set (70 PDB structures) (Wang et al. 2026)	0.052±0.057 F1 (attention; mean over structures) unitless · higher Uncertainty: SD 0.057 Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source RFamLlama (Large) zero-shot secondary structure on TS test set (70 PDB structures) (attention, F1 score, mean) lit-rna-20261011-rnalm-zeroshot-protocol-t4-f1-score-mean Aggregation: Not reported Zero-shot benchmarking of RNA language models: Supplementary Information; Zero-shot benchmarking of RNA language models in structural, functional, and evolutionary learning · Supplementary Table 4, row 'RFamLlama (Large)', column 'F1 score' (Mean ± Std)
Configuration: RiNALMo (giga) (zero-shot, Wang et al. 2026)	Protocol: Wang et al. 2026 zero-shot secondary structure, TS test set (70 PDB structures), attention, F1 score (mean over structures) Dataset: TS test set (70 PDB structures) (Wang et al. 2026)	0.414±0.255 F1 (attention; mean over structures) unitless · higher Uncertainty: SD 0.255 Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source RiNALMo (giga) zero-shot secondary structure on TS test set (70 PDB structures) (attention, F1 score, mean) lit-rna-20261011-rnalm-zeroshot-protocol-t4-f1-score-mean Aggregation: Not reported Zero-shot benchmarking of RNA language models: Supplementary Information; Zero-shot benchmarking of RNA language models in structural, functional, and evolutionary learning · Supplementary Table 4, row 'RiNALMo (giga)', column 'F1 score' (Mean ± Std)
Configuration: RiNALMo (mega) (zero-shot, Wang et al. 2026)	Protocol: Wang et al. 2026 zero-shot secondary structure, TS test set (70 PDB structures), attention, F1 score (mean over structures) Dataset: TS test set (70 PDB structures) (Wang et al. 2026)	0.342±0.225 F1 (attention; mean over structures) unitless · higher Uncertainty: SD 0.225 Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source RiNALMo (mega) zero-shot secondary structure on TS test set (70 PDB structures) (attention, F1 score, mean) lit-rna-20261011-rnalm-zeroshot-protocol-t4-f1-score-mean Aggregation: Not reported Zero-shot benchmarking of RNA language models: Supplementary Information; Zero-shot benchmarking of RNA language models in structural, functional, and evolutionary learning · Supplementary Table 4, row 'RiNALMo (mega)', column 'F1 score' (Mean ± Std)
Configuration: RiNALMo (micro) (zero-shot, Wang et al. 2026)	Protocol: Wang et al. 2026 zero-shot secondary structure, TS test set (70 PDB structures), attention, F1 score (mean over structures) Dataset: TS test set (70 PDB structures) (Wang et al. 2026)	0.268±0.184 F1 (attention; mean over structures) unitless · higher Uncertainty: SD 0.184 Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source RiNALMo (micro) zero-shot secondary structure on TS test set (70 PDB structures) (attention, F1 score, mean) lit-rna-20261011-rnalm-zeroshot-protocol-t4-f1-score-mean Aggregation: Not reported Zero-shot benchmarking of RNA language models: Supplementary Information; Zero-shot benchmarking of RNA language models in structural, functional, and evolutionary learning · Supplementary Table 4, row 'RiNALMo (micro)', column 'F1 score' (Mean ± Std)
Configuration: RNA-FM (zero-shot, Wang et al. 2026)	Protocol: Wang et al. 2026 zero-shot secondary structure, TS test set (70 PDB structures), attention, F1 score (mean over structures) Dataset: TS test set (70 PDB structures) (Wang et al. 2026)	0.267±0.239 F1 (attention; mean over structures) unitless · higher Uncertainty: SD 0.239 Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source RNA-FM zero-shot secondary structure on TS test set (70 PDB structures) (attention, F1 score, mean) lit-rna-20261011-rnalm-zeroshot-protocol-t4-f1-score-mean Aggregation: Not reported Zero-shot benchmarking of RNA language models: Supplementary Information; Zero-shot benchmarking of RNA language models in structural, functional, and evolutionary learning · Supplementary Table 4, row 'RNA-FM', column 'F1 score' (Mean ± Std)
Configuration: RNA-km (zero-shot, Wang et al. 2026)	Protocol: Wang et al. 2026 zero-shot secondary structure, TS test set (70 PDB structures), attention, F1 score (mean over structures) Dataset: TS test set (70 PDB structures) (Wang et al. 2026)	0.492±0.236 F1 (attention; mean over structures) unitless · higher Uncertainty: SD 0.236 Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source RNA-km zero-shot secondary structure on TS test set (70 PDB structures) (attention, F1 score, mean) lit-rna-20261011-rnalm-zeroshot-protocol-t4-f1-score-mean Aggregation: Not reported Zero-shot benchmarking of RNA language models: Supplementary Information; Zero-shot benchmarking of RNA language models in structural, functional, and evolutionary learning · Supplementary Table 4, row 'RNA-km', column 'F1 score' (Mean ± Std)
Configuration: RNA-MSM (zero-shot, Wang et al. 2026)	Protocol: Wang et al. 2026 zero-shot secondary structure, TS test set (70 PDB structures), attention, F1 score (mean over structures) Dataset: TS test set (70 PDB structures) (Wang et al. 2026)	0.582±0.217 F1 (attention; mean over structures) unitless · higher Uncertainty: SD 0.217 Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source RNA-MSM zero-shot secondary structure on TS test set (70 PDB structures) (attention, F1 score, mean) lit-rna-20261011-rnalm-zeroshot-protocol-t4-f1-score-mean Aggregation: Not reported Zero-shot benchmarking of RNA language models: Supplementary Information; Zero-shot benchmarking of RNA language models in structural, functional, and evolutionary learning · Supplementary Table 4, row 'RNA-MSM', column 'F1 score' (Mean ± Std)

Source checking is not independent reproduction. Release 2026-10-22-1b7cf027e4f6.

Methods and evaluation design

Procedure, tasks and evaluated configurations

Recorded evaluations

Each evaluation records what was tested and under which conditions.

Explore all linked results

Baseline coverage

Reference methods help show what a model adds beyond simple controls. We track a null control and a conventional method for each protocol.

0 of 2 active baseline roles have published Rewire measurements in this release. Measurements on a selected protocol do not establish coverage of an entire suite.

No execution recipe linked to this protocol. Recipe availability does not establish a completed evaluation.

Author-reported evaluations: 1
External evaluations: 31

Literature evidence is not a Rewire measurement. Executed but unpublished runs and private review status are not included.

Null control

Proposed control: requires review

Protocol-valid seeded random ranking; no fitting on assay labels

Protocol-specific applicability, permitted inputs, access, split, evaluator and execution requirements need review before implementation or execution.

This is a suggested selection rule, not a validated method or a measured score.

Conventional reference

Proposed control: requires review

Established label-free reference with explicitly permitted sequence/MSA inputs

Protocol-specific applicability, permitted inputs, access, split, evaluator and execution requirements need review before implementation or execution.

This is a suggested selection rule, not a validated method or a measured score.

Protocol coverage CSV (gzip) · Model evaluation matrix (gzip) · Source table (gzip) · Release and checksums (gzip)

Coverage is derived from release 2026-10-22-1b7cf027e4f6. Source citations describe the original records; they do not validate an unreviewed baseline proposal. No results have been generated by this audit.

Run instructions

No runnable recipe has been reviewed for this protocol. Dataset access, model requirements, licences and compute requirements must be checked against its sources before execution.

Strengths, limitations and unresolved questions

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.

0 evidence rows matching the loaded filters

Claims, original sources and review scope · Release 2026-10-22-1b7cf027e4f6
Property and statement	Original source and location	Review and provenance

No evidence rows match these filters. Choose another scope or clear the search.

How to read the evidence

Sources and history

Release 2026-10-22-1b7cf027e4f6 · Record review: source checked

2 source records and release history

Zero-shot benchmarking of RNA language models: Supplementary Information · Original source · bbag098_supplymentary_information_20260209.docx from the Europe PMC supplementary files archive for PMC12963973
Zero-shot benchmarking of RNA language models in structural, functional, and evolutionary learning · Original source · Briefings in Bioinformatics 2026, bbag098 (issue date 2026-03-01); PMC12963973 full-text XML

Download this release (gzip)

Technical metadata and extraction receipts

Stable ID: lit-rna-20261011-rnalm-zeroshot-protocol-t4-f1-score-mean

areas: rna-transcriptomes
contexts: research
protocol: Zero-shot: symmetrised, APC-corrected attention map from one head-layer and threshold chosen on VL1 by F1. Metric F1 score, mean over structures.
version: Wang et al. 2026
metric: f1-score
metric direction: higher
unit: unitless
metric qualifier: attention; mean over structures
source locator: Supplementary Table 4 caption and column 'F1 score'; main text Methods
limitations: Zero-shot only; no fine-tuning or task head.; Base pairs come from 3D structures; one head-layer and threshold per model chosen on VL1.

Related records

Suggest a correction