Protocol

FlexMS molecule-to-spectrum prediction on MassSpecGym, embedder-predictor grid (Table 2)

Predict a binned MS/MS spectrum from a molecule with each embedder-predictor combination; score cosine and Jensen-Shannon similarity to the observed spectrum.

33 evaluations · 66 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.

33 recorded evaluations, 66 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

33 evaluations · 66 results. Different protocols are not a single leaderboard.

Protocol: FlexMS molecule-to-spectrum prediction on MassSpecGym, embedder-predictor grid (Table 2) · Dataset: MassSpecGym (FlexMS spectrum prediction)

Sorted by Jensen shannon similarity (test split; bin resolution not stated in the caption) (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	Jensen shannon similarity (test split; bin resolution not stated in the caption)	Spectrum cosine similarity (test split; bin resolution not stated in the caption)
MassFormer predictor with GraphormerV2 embedder (FlexMS)Evaluation	0.516 (best in column)	0.357 (best in column)
MassFormer predictor with GIN embedder (FlexMS)Evaluation	0.494	0.329
MassFormer predictor with All-Descriptors MLP embedder (FlexMS)Evaluation	0.491	0.332
MassFormer predictor with AttentiveFP embedder (FlexMS)Evaluation	0.489	0.323
NEIMS predictor with GraphormerV2 embedder (FlexMS)Evaluation	0.488	0.354
MassFormer predictor with GCN embedder (FlexMS)Evaluation	0.484	0.306
MassFormer predictor with GAT embedder (FlexMS)Evaluation	0.483	0.317
MassFormer predictor with Daylight fingerprint embedder (FlexMS)Evaluation	0.481	0.324
MassFormer predictor with ChemBERTa embedder (FlexMS)Evaluation	0.48	0.317
MassFormer predictor with CNN embedder (FlexMS)Evaluation	0.477	0.316
MassFormer predictor with ESPF embedder (FlexMS)Evaluation	0.475	0.31
MassFormer predictor with Transformer embedder (FlexMS)Evaluation	0.471	0.31
NEIMS predictor with All-Descriptors MLP embedder (FlexMS)Evaluation	0.47	0.319
NEIMS predictor with GCN embedder (FlexMS)Evaluation	0.468	0.302
NEIMS predictor with Daylight fingerprint embedder (FlexMS)Evaluation	0.462	0.308
NEIMS predictor with AttentiveFP embedder (FlexMS)Evaluation	0.461	0.314
NEIMS predictor with GIN embedder (FlexMS)Evaluation	0.461	0.324
NEIMS predictor with ChemBERTa embedder (FlexMS)Evaluation	0.46	0.303
NEIMS predictor with GAT embedder (FlexMS)Evaluation	0.458	0.304
MolMS predictor with GraphormerV2 embedder (FlexMS)Evaluation	0.457	0.271
NEIMS predictor with Transformer embedder (FlexMS)Evaluation	0.456	0.305
NEIMS predictor with CNN embedder (FlexMS)Evaluation	0.455	0.303
NEIMS predictor with ESPF embedder (FlexMS)Evaluation	0.454	0.297
MolMS predictor with All-Descriptors MLP embedder (FlexMS)Evaluation	0.446	0.241
MolMS predictor with AttentiveFP embedder (FlexMS)Evaluation	0.442	0.248
MolMS predictor with GIN embedder (FlexMS)Evaluation	0.442	0.247
MolMS predictor with GCN embedder (FlexMS)Evaluation	0.44	0.24
MolMS predictor with GAT embedder (FlexMS)Evaluation	0.439	0.24
MolMS predictor with ChemBERTa embedder (FlexMS)Evaluation	0.436	0.233
MolMS predictor with Daylight fingerprint embedder (FlexMS)Evaluation	0.436	0.229
MolMS predictor with CNN embedder (FlexMS)Evaluation	0.433	0.226
MolMS predictor with ESPF embedder (FlexMS)Evaluation	0.43	0.212
MolMS predictor with Transformer embedder (FlexMS)Evaluation	0.43	0.224

All 66 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: MassFormer predictor with All-Descriptors MLP embedder (FlexMS)	Protocol: FlexMS molecule-to-spectrum prediction on MassSpecGym, embedder-predictor grid (Table 2) Dataset: MassSpecGym (FlexMS spectrum prediction)	0.491 Jensen shannon similarity (test split; bin resolution not stated in the caption) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MassFormer with All-Descriptors MLP on MassSpecGym spectrum prediction lit20261011-zhong2026-protocol-t2-massspecgym Aggregation: Not reported FlexMS: A Unified Public Benchmark for Molecule Tandem Mass Spectrum Prediction · Table 2 section 'Jensen-Shannon Similarity (JS Sim)', row 'Mass', column 'AllDes'
Configuration: MassFormer predictor with All-Descriptors MLP embedder (FlexMS)	Protocol: FlexMS molecule-to-spectrum prediction on MassSpecGym, embedder-predictor grid (Table 2) Dataset: MassSpecGym (FlexMS spectrum prediction)	0.332 Spectrum cosine similarity (test split; bin resolution not stated in the caption) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MassFormer with All-Descriptors MLP on MassSpecGym spectrum prediction lit20261011-zhong2026-protocol-t2-massspecgym Aggregation: Not reported FlexMS: A Unified Public Benchmark for Molecule Tandem Mass Spectrum Prediction · Table 2 section 'Cosine Similarity (Cos Sim)', row 'Mass', column 'AllDes'
Configuration: MassFormer predictor with AttentiveFP embedder (FlexMS)	Protocol: FlexMS molecule-to-spectrum prediction on MassSpecGym, embedder-predictor grid (Table 2) Dataset: MassSpecGym (FlexMS spectrum prediction)	0.489 Jensen shannon similarity (test split; bin resolution not stated in the caption) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MassFormer with AttentiveFP on MassSpecGym spectrum prediction lit20261011-zhong2026-protocol-t2-massspecgym Aggregation: Not reported FlexMS: A Unified Public Benchmark for Molecule Tandem Mass Spectrum Prediction · Table 2 section 'Jensen-Shannon Similarity (JS Sim)', row 'Mass', column 'Attn'
Configuration: MassFormer predictor with AttentiveFP embedder (FlexMS)	Protocol: FlexMS molecule-to-spectrum prediction on MassSpecGym, embedder-predictor grid (Table 2) Dataset: MassSpecGym (FlexMS spectrum prediction)	0.323 Spectrum cosine similarity (test split; bin resolution not stated in the caption) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MassFormer with AttentiveFP on MassSpecGym spectrum prediction lit20261011-zhong2026-protocol-t2-massspecgym Aggregation: Not reported FlexMS: A Unified Public Benchmark for Molecule Tandem Mass Spectrum Prediction · Table 2 section 'Cosine Similarity (Cos Sim)', row 'Mass', column 'Attn'
Configuration: MassFormer predictor with ChemBERTa embedder (FlexMS)	Protocol: FlexMS molecule-to-spectrum prediction on MassSpecGym, embedder-predictor grid (Table 2) Dataset: MassSpecGym (FlexMS spectrum prediction)	0.48 Jensen shannon similarity (test split; bin resolution not stated in the caption) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MassFormer with ChemBERTa on MassSpecGym spectrum prediction lit20261011-zhong2026-protocol-t2-massspecgym Aggregation: Not reported FlexMS: A Unified Public Benchmark for Molecule Tandem Mass Spectrum Prediction · Table 2 section 'Jensen-Shannon Similarity (JS Sim)', row 'Mass', column 'CBTa'
Configuration: MassFormer predictor with ChemBERTa embedder (FlexMS)	Protocol: FlexMS molecule-to-spectrum prediction on MassSpecGym, embedder-predictor grid (Table 2) Dataset: MassSpecGym (FlexMS spectrum prediction)	0.317 Spectrum cosine similarity (test split; bin resolution not stated in the caption) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MassFormer with ChemBERTa on MassSpecGym spectrum prediction lit20261011-zhong2026-protocol-t2-massspecgym Aggregation: Not reported FlexMS: A Unified Public Benchmark for Molecule Tandem Mass Spectrum Prediction · Table 2 section 'Cosine Similarity (Cos Sim)', row 'Mass', column 'CBTa'
Configuration: MassFormer predictor with CNN embedder (FlexMS)	Protocol: FlexMS molecule-to-spectrum prediction on MassSpecGym, embedder-predictor grid (Table 2) Dataset: MassSpecGym (FlexMS spectrum prediction)	0.477 Jensen shannon similarity (test split; bin resolution not stated in the caption) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MassFormer with CNN on MassSpecGym spectrum prediction lit20261011-zhong2026-protocol-t2-massspecgym Aggregation: Not reported FlexMS: A Unified Public Benchmark for Molecule Tandem Mass Spectrum Prediction · Table 2 section 'Jensen-Shannon Similarity (JS Sim)', row 'Mass', column 'CNN'
Configuration: MassFormer predictor with CNN embedder (FlexMS)	Protocol: FlexMS molecule-to-spectrum prediction on MassSpecGym, embedder-predictor grid (Table 2) Dataset: MassSpecGym (FlexMS spectrum prediction)	0.316 Spectrum cosine similarity (test split; bin resolution not stated in the caption) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MassFormer with CNN on MassSpecGym spectrum prediction lit20261011-zhong2026-protocol-t2-massspecgym Aggregation: Not reported FlexMS: A Unified Public Benchmark for Molecule Tandem Mass Spectrum Prediction · Table 2 section 'Cosine Similarity (Cos Sim)', row 'Mass', column 'CNN'
Configuration: MassFormer predictor with Daylight fingerprint embedder (FlexMS)	Protocol: FlexMS molecule-to-spectrum prediction on MassSpecGym, embedder-predictor grid (Table 2) Dataset: MassSpecGym (FlexMS spectrum prediction)	0.481 Jensen shannon similarity (test split; bin resolution not stated in the caption) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MassFormer with Daylight fingerprint on MassSpecGym spectrum prediction lit20261011-zhong2026-protocol-t2-massspecgym Aggregation: Not reported FlexMS: A Unified Public Benchmark for Molecule Tandem Mass Spectrum Prediction · Table 2 section 'Jensen-Shannon Similarity (JS Sim)', row 'Mass', column 'DL'
Configuration: MassFormer predictor with Daylight fingerprint embedder (FlexMS)	Protocol: FlexMS molecule-to-spectrum prediction on MassSpecGym, embedder-predictor grid (Table 2) Dataset: MassSpecGym (FlexMS spectrum prediction)	0.324 Spectrum cosine similarity (test split; bin resolution not stated in the caption) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MassFormer with Daylight fingerprint on MassSpecGym spectrum prediction lit20261011-zhong2026-protocol-t2-massspecgym Aggregation: Not reported FlexMS: A Unified Public Benchmark for Molecule Tandem Mass Spectrum Prediction · Table 2 section 'Cosine Similarity (Cos Sim)', row 'Mass', column 'DL'
Configuration: MassFormer predictor with ESPF embedder (FlexMS)	Protocol: FlexMS molecule-to-spectrum prediction on MassSpecGym, embedder-predictor grid (Table 2) Dataset: MassSpecGym (FlexMS spectrum prediction)	0.475 Jensen shannon similarity (test split; bin resolution not stated in the caption) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MassFormer with ESPF on MassSpecGym spectrum prediction lit20261011-zhong2026-protocol-t2-massspecgym Aggregation: Not reported FlexMS: A Unified Public Benchmark for Molecule Tandem Mass Spectrum Prediction · Table 2 section 'Jensen-Shannon Similarity (JS Sim)', row 'Mass', column 'ESPF'
Configuration: MassFormer predictor with ESPF embedder (FlexMS)	Protocol: FlexMS molecule-to-spectrum prediction on MassSpecGym, embedder-predictor grid (Table 2) Dataset: MassSpecGym (FlexMS spectrum prediction)	0.31 Spectrum cosine similarity (test split; bin resolution not stated in the caption) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MassFormer with ESPF on MassSpecGym spectrum prediction lit20261011-zhong2026-protocol-t2-massspecgym Aggregation: Not reported FlexMS: A Unified Public Benchmark for Molecule Tandem Mass Spectrum Prediction · Table 2 section 'Cosine Similarity (Cos Sim)', row 'Mass', column 'ESPF'
Configuration: MassFormer predictor with GAT embedder (FlexMS)	Protocol: FlexMS molecule-to-spectrum prediction on MassSpecGym, embedder-predictor grid (Table 2) Dataset: MassSpecGym (FlexMS spectrum prediction)	0.483 Jensen shannon similarity (test split; bin resolution not stated in the caption) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MassFormer with GAT on MassSpecGym spectrum prediction lit20261011-zhong2026-protocol-t2-massspecgym Aggregation: Not reported FlexMS: A Unified Public Benchmark for Molecule Tandem Mass Spectrum Prediction · Table 2 section 'Jensen-Shannon Similarity (JS Sim)', row 'Mass', column 'GAT'
Configuration: MassFormer predictor with GAT embedder (FlexMS)	Protocol: FlexMS molecule-to-spectrum prediction on MassSpecGym, embedder-predictor grid (Table 2) Dataset: MassSpecGym (FlexMS spectrum prediction)	0.317 Spectrum cosine similarity (test split; bin resolution not stated in the caption) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MassFormer with GAT on MassSpecGym spectrum prediction lit20261011-zhong2026-protocol-t2-massspecgym Aggregation: Not reported FlexMS: A Unified Public Benchmark for Molecule Tandem Mass Spectrum Prediction · Table 2 section 'Cosine Similarity (Cos Sim)', row 'Mass', column 'GAT'
Configuration: MassFormer predictor with GCN embedder (FlexMS)	Protocol: FlexMS molecule-to-spectrum prediction on MassSpecGym, embedder-predictor grid (Table 2) Dataset: MassSpecGym (FlexMS spectrum prediction)	0.484 Jensen shannon similarity (test split; bin resolution not stated in the caption) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MassFormer with GCN on MassSpecGym spectrum prediction lit20261011-zhong2026-protocol-t2-massspecgym Aggregation: Not reported FlexMS: A Unified Public Benchmark for Molecule Tandem Mass Spectrum Prediction · Table 2 section 'Jensen-Shannon Similarity (JS Sim)', row 'Mass', column 'GCN'
Configuration: MassFormer predictor with GCN embedder (FlexMS)	Protocol: FlexMS molecule-to-spectrum prediction on MassSpecGym, embedder-predictor grid (Table 2) Dataset: MassSpecGym (FlexMS spectrum prediction)	0.306 Spectrum cosine similarity (test split; bin resolution not stated in the caption) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MassFormer with GCN on MassSpecGym spectrum prediction lit20261011-zhong2026-protocol-t2-massspecgym Aggregation: Not reported FlexMS: A Unified Public Benchmark for Molecule Tandem Mass Spectrum Prediction · Table 2 section 'Cosine Similarity (Cos Sim)', row 'Mass', column 'GCN'
Configuration: MassFormer predictor with GraphormerV2 embedder (FlexMS)	Protocol: FlexMS molecule-to-spectrum prediction on MassSpecGym, embedder-predictor grid (Table 2) Dataset: MassSpecGym (FlexMS spectrum prediction)	0.516 Jensen shannon similarity (test split; bin resolution not stated in the caption) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MassFormer with GraphormerV2 on MassSpecGym spectrum prediction lit20261011-zhong2026-protocol-t2-massspecgym Aggregation: Not reported FlexMS: A Unified Public Benchmark for Molecule Tandem Mass Spectrum Prediction · Table 2 section 'Jensen-Shannon Similarity (JS Sim)', row 'Mass', column 'GFv2'
Configuration: MassFormer predictor with GraphormerV2 embedder (FlexMS)	Protocol: FlexMS molecule-to-spectrum prediction on MassSpecGym, embedder-predictor grid (Table 2) Dataset: MassSpecGym (FlexMS spectrum prediction)	0.357 Spectrum cosine similarity (test split; bin resolution not stated in the caption) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MassFormer with GraphormerV2 on MassSpecGym spectrum prediction lit20261011-zhong2026-protocol-t2-massspecgym Aggregation: Not reported FlexMS: A Unified Public Benchmark for Molecule Tandem Mass Spectrum Prediction · Table 2 section 'Cosine Similarity (Cos Sim)', row 'Mass', column 'GFv2'
Configuration: MassFormer predictor with GIN embedder (FlexMS)	Protocol: FlexMS molecule-to-spectrum prediction on MassSpecGym, embedder-predictor grid (Table 2) Dataset: MassSpecGym (FlexMS spectrum prediction)	0.494 Jensen shannon similarity (test split; bin resolution not stated in the caption) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MassFormer with GIN on MassSpecGym spectrum prediction lit20261011-zhong2026-protocol-t2-massspecgym Aggregation: Not reported FlexMS: A Unified Public Benchmark for Molecule Tandem Mass Spectrum Prediction · Table 2 section 'Jensen-Shannon Similarity (JS Sim)', row 'Mass', column 'GIN'
Configuration: MassFormer predictor with GIN embedder (FlexMS)	Protocol: FlexMS molecule-to-spectrum prediction on MassSpecGym, embedder-predictor grid (Table 2) Dataset: MassSpecGym (FlexMS spectrum prediction)	0.329 Spectrum cosine similarity (test split; bin resolution not stated in the caption) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MassFormer with GIN on MassSpecGym spectrum prediction lit20261011-zhong2026-protocol-t2-massspecgym Aggregation: Not reported FlexMS: A Unified Public Benchmark for Molecule Tandem Mass Spectrum Prediction · Table 2 section 'Cosine Similarity (Cos Sim)', row 'Mass', column 'GIN'
Configuration: MassFormer predictor with Transformer embedder (FlexMS)	Protocol: FlexMS molecule-to-spectrum prediction on MassSpecGym, embedder-predictor grid (Table 2) Dataset: MassSpecGym (FlexMS spectrum prediction)	0.471 Jensen shannon similarity (test split; bin resolution not stated in the caption) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MassFormer with Transformer on MassSpecGym spectrum prediction lit20261011-zhong2026-protocol-t2-massspecgym Aggregation: Not reported FlexMS: A Unified Public Benchmark for Molecule Tandem Mass Spectrum Prediction · Table 2 section 'Jensen-Shannon Similarity (JS Sim)', row 'Mass', column 'Trans'
Configuration: MassFormer predictor with Transformer embedder (FlexMS)	Protocol: FlexMS molecule-to-spectrum prediction on MassSpecGym, embedder-predictor grid (Table 2) Dataset: MassSpecGym (FlexMS spectrum prediction)	0.31 Spectrum cosine similarity (test split; bin resolution not stated in the caption) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MassFormer with Transformer on MassSpecGym spectrum prediction lit20261011-zhong2026-protocol-t2-massspecgym Aggregation: Not reported FlexMS: A Unified Public Benchmark for Molecule Tandem Mass Spectrum Prediction · Table 2 section 'Cosine Similarity (Cos Sim)', row 'Mass', column 'Trans'
Configuration: MolMS predictor with All-Descriptors MLP embedder (FlexMS)	Protocol: FlexMS molecule-to-spectrum prediction on MassSpecGym, embedder-predictor grid (Table 2) Dataset: MassSpecGym (FlexMS spectrum prediction)	0.446 Jensen shannon similarity (test split; bin resolution not stated in the caption) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MolMS with All-Descriptors MLP on MassSpecGym spectrum prediction lit20261011-zhong2026-protocol-t2-massspecgym Aggregation: Not reported FlexMS: A Unified Public Benchmark for Molecule Tandem Mass Spectrum Prediction · Table 2 section 'Jensen-Shannon Similarity (JS Sim)', row 'MolMS', column 'AllDes'
Configuration: MolMS predictor with All-Descriptors MLP embedder (FlexMS)	Protocol: FlexMS molecule-to-spectrum prediction on MassSpecGym, embedder-predictor grid (Table 2) Dataset: MassSpecGym (FlexMS spectrum prediction)	0.241 Spectrum cosine similarity (test split; bin resolution not stated in the caption) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MolMS with All-Descriptors MLP on MassSpecGym spectrum prediction lit20261011-zhong2026-protocol-t2-massspecgym Aggregation: Not reported FlexMS: A Unified Public Benchmark for Molecule Tandem Mass Spectrum Prediction · Table 2 section 'Cosine Similarity (Cos Sim)', row 'MolMS', column 'AllDes'
Configuration: MolMS predictor with AttentiveFP embedder (FlexMS)	Protocol: FlexMS molecule-to-spectrum prediction on MassSpecGym, embedder-predictor grid (Table 2) Dataset: MassSpecGym (FlexMS spectrum prediction)	0.442 Jensen shannon similarity (test split; bin resolution not stated in the caption) unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Independent external evaluation · Source checked Methods, coverage and source MolMS with AttentiveFP on MassSpecGym spectrum prediction lit20261011-zhong2026-protocol-t2-massspecgym Aggregation: Not reported FlexMS: A Unified Public Benchmark for Molecule Tandem Mass Spectrum Prediction · Table 2 section 'Jensen-Shannon Similarity (JS Sim)', row 'MolMS', column 'Attn'

Source checking is not independent reproduction. Release 2026-10-23-5347695500ed.

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.

External evaluations: 33

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

Null control

Proposed control: requires review

Select a task-valid null control after reviewing inputs and metric

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

Select an upstream conventional reference after reviewing the full protocol

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 2026-10-23-5347695500ed. 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.

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 2026-10-23-5347695500ed
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 2026-10-23-5347695500ed · Record review: source checked

1 source record and release history

FlexMS: A Unified Public Benchmark for Molecule Tandem Mass Spectrum Prediction · Original source · arXiv 2602.22822v4 (v1 submitted 2026-02-26; v4 dated 2026-06-24 on the PDF)

Download this release (gzip)

Technical metadata and extraction receipts

Stable ID: lit20261011-zhong2026-protocol-t2-massspecgym

areas: metabolomics
contexts: research
protocol: Embedder and predictor trained end to end with a cosine loss, AdamW, up to 100 epochs with early stopping on validation cosine, batch 512, seed 42 (Appendix C).
version: Table 2
task: Molecule-to-MS/MS spectrum prediction
source locator: Table 2 caption; Section 4.2; Appendix C
limitations: The Table 2 caption does not state the bin resolution; Section 4.2 refers to several resolutions.; Single seed (42); no uncertainty reported.; Bold, double-line and underline rank marks are not recorded.
missing metadata: metric implementation: reason: unreported; note: Bin width for Table 2 not stated in the caption

Related records

uses data: MassSpecGym (FlexMS spectrum prediction)
evaluates task: MassSpecGym Spectrum simulation
assessment: MassFormer with All-Descriptors MLP on MassSpecGym spectrum prediction
assessment: MassFormer with AttentiveFP on MassSpecGym spectrum prediction
assessment: MassFormer with ChemBERTa on MassSpecGym spectrum prediction
assessment: MassFormer with CNN on MassSpecGym spectrum prediction
assessment: MassFormer with Daylight fingerprint on MassSpecGym spectrum prediction
assessment: MassFormer with ESPF on MassSpecGym spectrum prediction
assessment: MassFormer with GAT on MassSpecGym spectrum prediction
assessment: MassFormer with GCN on MassSpecGym spectrum prediction
assessment: MassFormer with GraphormerV2 on MassSpecGym spectrum prediction
assessment: MassFormer with GIN on MassSpecGym spectrum prediction
assessment: MassFormer with Transformer on MassSpecGym spectrum prediction
assessment: MolMS with All-Descriptors MLP on MassSpecGym spectrum prediction
assessment: MolMS with AttentiveFP on MassSpecGym spectrum prediction
assessment: MolMS with ChemBERTa on MassSpecGym spectrum prediction
assessment: MolMS with CNN on MassSpecGym spectrum prediction
assessment: MolMS with Daylight fingerprint on MassSpecGym spectrum prediction
assessment: MolMS with ESPF on MassSpecGym spectrum prediction
assessment: MolMS with GAT on MassSpecGym spectrum prediction
assessment: MolMS with GCN on MassSpecGym spectrum prediction
assessment: MolMS with GraphormerV2 on MassSpecGym spectrum prediction
assessment: MolMS with GIN on MassSpecGym spectrum prediction
assessment: MolMS with Transformer on MassSpecGym spectrum prediction
assessment: NEIMS with All-Descriptors MLP on MassSpecGym spectrum prediction
assessment: NEIMS with AttentiveFP on MassSpecGym spectrum prediction
assessment: NEIMS with ChemBERTa on MassSpecGym spectrum prediction
assessment: NEIMS with CNN on MassSpecGym spectrum prediction
assessment: NEIMS with Daylight fingerprint on MassSpecGym spectrum prediction
assessment: NEIMS with ESPF on MassSpecGym spectrum prediction
assessment: NEIMS with GAT on MassSpecGym spectrum prediction
assessment: NEIMS with GCN on MassSpecGym spectrum prediction
assessment: NEIMS with GraphormerV2 on MassSpecGym spectrum prediction
assessment: NEIMS with GIN on MassSpecGym spectrum prediction
assessment: NEIMS with Transformer on MassSpecGym spectrum prediction

Suggest a correction