ChemFormer two-stage retrieval, original candidate canonicalization (MassSpecGym in the Wild)
ChemFormer two-stage retrieval, original candidate canonicalization; no settings beyond its name are recorded.
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.
Evaluations and results
4 evaluations · 4 results. Different protocols are not a single leaderboard.
Dataset: MassSpecGym test fold, formula-known retrieval candidates (bonus) (MassSpecGym in the Wild)
Sorted by Top-1 accuracy (retrieval) (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.
All 4 result rows with coverage, uncertainty and sources
Filter evaluations
Applied filters: All linked evaluations
| Tested configuration | Protocol and dataset | Finding | Evidence and details |
|---|---|---|---|
| Configuration: ChemFormer two-stage retrieval, original candidate canonicalization (MassSpecGym in the Wild) | Protocol: MassSpecGym formula-known retrieval (bonus), Hit rate@1 (MassSpecGym in the Wild Table 3) Dataset: MassSpecGym test fold, formula-known retrieval candidates (bonus) (MassSpecGym in the Wild) | 63.89 (62.71-65.03)% Top-1 accuracy (retrieval) percent · higher Uncertainty: 99.9% CI 62.71 to 65.03. Range printed in parentheses after the mean. Appendix 'Confidence Intervals': 99.9% bootstrap interval over 20,000 resamples of the test set. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourcelit-otheromics-20261011-msgwild-protocol-formula-retrieval-hit-rate-1 Aggregation: Not reported MassSpecGym in the Wild: Uncovering and Correcting Evaluation Pitfalls in AI-Driven Molecule Discovery · Table 3, section 'Exploiting Canonicalization Artifacts (Not Valid Under Corrected Candidate Canonicalization)', row '(2) ChemFormer Two-Stage Retrieval (inflated)', column 'Hit rate@1' |
| Configuration: ChemFormer two-stage retrieval, original candidate canonicalization (MassSpecGym in the Wild) | Protocol: MassSpecGym formula-known retrieval (bonus), Hit rate@20 (MassSpecGym in the Wild Table 3) Dataset: MassSpecGym test fold, formula-known retrieval candidates (bonus) (MassSpecGym in the Wild) | 75.86 (74.79-76.89)% Top-20 accuracy (retrieval) percent · higher Uncertainty: 99.9% CI 74.79 to 76.89. Range printed in parentheses after the mean. Appendix 'Confidence Intervals': 99.9% bootstrap interval over 20,000 resamples of the test set. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourcelit-otheromics-20261011-msgwild-protocol-formula-retrieval-hit-rate-20 Aggregation: Not reported MassSpecGym in the Wild: Uncovering and Correcting Evaluation Pitfalls in AI-Driven Molecule Discovery · Table 3, section 'Exploiting Canonicalization Artifacts (Not Valid Under Corrected Candidate Canonicalization)', row '(2) ChemFormer Two-Stage Retrieval (inflated)', column 'Hit rate@20' |
| Configuration: ChemFormer two-stage retrieval, original candidate canonicalization (MassSpecGym in the Wild) | Protocol: MassSpecGym formula-known retrieval (bonus), Hit rate@5 (MassSpecGym in the Wild Table 3) Dataset: MassSpecGym test fold, formula-known retrieval candidates (bonus) (MassSpecGym in the Wild) | 70.52 (69.37-71.70)% Top-5 accuracy (retrieval) percent · higher Uncertainty: 99.9% CI 69.37 to 71.70. Range printed in parentheses after the mean. Appendix 'Confidence Intervals': 99.9% bootstrap interval over 20,000 resamples of the test set. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourcelit-otheromics-20261011-msgwild-protocol-formula-retrieval-hit-rate-5 Aggregation: Not reported MassSpecGym in the Wild: Uncovering and Correcting Evaluation Pitfalls in AI-Driven Molecule Discovery · Table 3, section 'Exploiting Canonicalization Artifacts (Not Valid Under Corrected Candidate Canonicalization)', row '(2) ChemFormer Two-Stage Retrieval (inflated)', column 'Hit rate@5' |
| Configuration: ChemFormer two-stage retrieval, original candidate canonicalization (MassSpecGym in the Wild) | Protocol: MassSpecGym formula-known retrieval (bonus), MCES@1 (MassSpecGym in the Wild Table 3) Dataset: MassSpecGym test fold, formula-known retrieval candidates (bonus) (MassSpecGym in the Wild) | 5.39 (5.20-5.60) Top-1 MCES (retrieval) unitless · lower Uncertainty: 99.9% CI 5.20 to 5.60. Range printed in parentheses after the mean. Appendix 'Confidence Intervals': 99.9% bootstrap interval over 20,000 resamples of the test set. Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourcelit-otheromics-20261011-msgwild-protocol-formula-retrieval-mces-1 Aggregation: Not reported MassSpecGym in the Wild: Uncovering and Correcting Evaluation Pitfalls in AI-Driven Molecule Discovery · Table 3, section 'Exploiting Canonicalization Artifacts (Not Valid Under Corrected Candidate Canonicalization)', row '(2) ChemFormer Two-Stage Retrieval (inflated)', column 'MCES@1' |
Source checking is not independent reproduction. Release 2026-10-23-5347695500ed.
Use this configuration
How it works, versions and access
Additional methodological details, versions and access terms have not yet been extracted into a structured profile for this configuration. Consult the linked sources and evaluation records below.
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
| Property and statement | Original source and location | Review and provenance |
|---|
No evidence rows match these filters. Choose another scope or clear the search.
Sources and history
Release 2026-10-23-5347695500ed · Record review: source checked
1 source record and release history
- MassSpecGym in the Wild: Uncovering and Correcting Evaluation Pitfalls in AI-Driven Molecule Discovery · Original source · arXiv:2606.19624v1 (submitted 2026-06-17), arXiv HTML rendering
Technical metadata and extraction receipts
Stable ID: lit-otheromics-20261011-msgwild-config-chemformer-two-stage-retrieval-original-candidate-canonicalization
- areas
- metabolomics
- contexts
- research
- method types
- foundation_model
- reported name
- (2) ChemFormer Two-Stage Retrieval (inflated)
- foundation model eligible
- true
- parameters
- ChemFormer two-stage retrieval, original candidate canonicalization
- access reuse
- Not stated in the source; see the cited work or https://github.com/pluskal-lab/MassSpecGym
- source locator
- Table 3 row '(2) ChemFormer Two-Stage Retrieval (inflated)'
- missing metadata
- version: reason: unreported; note: Only the printed row label identifies the setup
- limitations
- Grey: exploits the SMILES canonicalization artifact. Source marks this row in grey.
- provenance note
- ChemFormer two-stage retrieval, original candidate canonicalization, as listed in MassSpecGym in the Wild Table 3.
Related records
- configuration of: ChemFormer
- system: ChemFormer two-stage retrieval, original candidate canonicalization on MassSpecGym formula-retrieval (Hit rate@1)
- system: ChemFormer two-stage retrieval, original candidate canonicalization on MassSpecGym formula-retrieval (Hit rate@20)
- system: ChemFormer two-stage retrieval, original candidate canonicalization on MassSpecGym formula-retrieval (Hit rate@5)
- system: ChemFormer two-stage retrieval, original candidate canonicalization on MassSpecGym formula-retrieval (MCES@1)