Research map / Probabilistic and scientific modeling / Molecules and proteins
Tartarus: A Benchmarking Platform for Realistic And Practical Inverse Molecular Design
AkshatKumar Nigam, Robert Pollice, Gary Tom, Kjell Jorner, John Willes, Luca Thiede, Anshul Kundaje, Alan Aspuru-Guzik
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 |
|---|---|---|---|
| use SELFIES instead of SMILES representation (ablation) | VAE · Protein ligand 1SYH | success rate: 12.3% -> 34.8% | Improved |
| use SELFIES instead of SMILES representation (ablation) | VAE · Protein ligand 1SYH QuickVina2 | docking score: -10.0 -> -10.4 | Improved |
| use SELFIES instead of SMILES representation (ablation) | LSTM-HC · Organic emitters multi-objective | objective: -0.24 -> -0.24 | No clear effect |
| use SELFIES instead of SMILES representation (ablation) | LSTM-HC · OPV PCE_PCDTBT - SAscore | objective: 31.79 -> 30.71 | Hurt |
| use SELFIES instead of SMILES representation (ablation) | VAE · OPV PCE_PCDTBT - SAscore | objective: 10.23 -> 29.24 | Improved |
| use SELFIES instead of SMILES representation (ablation) | LSTM-HC · SMILES vs SELFIES training time | training time | No clear effect |
| GA (GB-GA, JANUS) instead of VAE generative models | Reaction substrate benchmark · Reaction substrate activation energy | activation energy: 76.81 -> 47.56 | Improved |
| VAE models versus dataset best | SMILES-VAE · Reaction substrate activation energy | activation energy: 64.94 -> 76.81 | Hurt |
Closest papers by meaning
- Beyond Drug Discovery: The Nanotechnology Molecular Optimization (NMO) Benchmark (NeurIPS 2026 · similarity 0.87)
- MADE: Benchmark Environments for Closed-Loop Materials Discovery (ICML 2026 · similarity 0.86)
- Projecting Molecules into Synthesizable Chemical Spaces (ICML 2024 · similarity 0.85)
- Representing Molecules as Random Walks Over Interpretable Grammars (ICML 2024 · similarity 0.84)
- UMA: A Family of Universal Models for Atoms (NeurIPS 2025 · similarity 0.84)
- MANDREL: Modular Reinforcement Learning Pipelines for Material Discovery (AAAI 2024 · similarity 0.84)
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.
