Research map / Probabilistic and scientific modeling / Molecules and proteins
Prefix-Tree Decoding for Predicting Mass Spectra from Molecules
Samuel Goldman, John Bradshaw, Jiayi Xin, Connor W. Coley
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Predict fragment formulae and then assign their intensities | SCARF · NIST20 | Cosine similarity: 0.694 -> 0.726 | Improved |
| Predict fragment formulae and then assign their intensities | SCARF · NPLIB1 | Cosine similarity: 0.568 -> 0.536 | Hurt |
| Combine forward count and atom-loss difference predictions (ablation) | SCARF-Thread · NIST20, Coverage @ 300 | Coverage: 0.839 -> 0.907; 0.855 -> 0.907 | Improved |
| Combine forward count and atom-loss difference predictions (ablation) | SCARF-Thread · NPLIB1, Coverage @ 300 | Coverage: 0.681 -> 0.724; 0.708 -> 0.724 | Improved |
| Constrain predicted peaks to valid subformulae | SCARF · NIST20 | Valid: 0.947 -> 1.000 | Improved |
| Use SCARF rather than CFM-ID for spectrum prediction | NIST20 · 100 molecules | Time (s): 1114.7 -> 21.1 | Improved |
| Use SCARF predicted spectra for retrieval | NIST20 · Retrieval top 1 | Accuracy: 0.175 ± 0.003 -> 0.187 ± 0.004 | Improved |
| Use SCARF predicted spectra for retrieval | NIST20 · Retrieval top 5 | Accuracy: 0.522 ± 0.006 -> 0.541 ± 0.005 | Improved |
| Use SCARF rather than FixedVocab | NPLIB1 · Retrieval top 1 | Accuracy: 0.193 ± 0.003 -> 0.135 ± 0.007 | Hurt |
Closest papers by meaning
- DiffMS: Diffusion Generation of Molecules Conditioned on Mass Spectra (ICML 2025 · similarity 0.88)
- GLACIER: Rethinking Mass Spectrum Prediction as an Object Detection Problem (NeurIPS 2026 · similarity 0.87)
- Unraveling Molecular Structure: A Multimodal Spectroscopic Dataset for Chemistry (NeurIPS 2024 · similarity 0.86)
- MS-BART: Unified Modeling of Mass Spectra and Molecules for Structure Elucidation (NeurIPS 2025 · similarity 0.86)
- MassSpecGym: A benchmark for the discovery and identification of molecules (NeurIPS 2024 · similarity 0.86)
- MADGEN: Mass-Spec attends to De Novo Molecular generation (ICLR 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
