Research map / Probabilistic and scientific modeling / Molecules and proteins
Bootstrapped Training of Score-Conditioned Generator for Offline Design of Biological Sequences
Minsu Kim, Federico Berto, Sungsoo Ahn, Jinkyoo Park
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 |
|---|---|---|---|
| Add rank-based reweighting (ablation) | BootGen · RNA-A | average score: 0.388 -> 0.483 | Improved |
| Add rank-based reweighting (ablation) | BootGen · GFP | average score: 0.457 -> 0.281 | Hurt |
| Add bootstrapping to rank-based reweighting (ablation) | BootGen · UTR | average score: 0.586 -> 0.689 | Improved |
| Add bootstrapping to rank-based reweighting (no filtering) (ablation) | BootGen · RNA-A | average score: 0.483 -> 0.408 | Hurt |
| Add filtering to rank-based reweighting (ablation) | BootGen · GFP | average score: 0.281 -> 0.783 | Improved |
| Add filtering to rank-based reweighting (ablation) | BootGen · TFbind8 | average score: 0.662 -> 0.833 | Improved |
| Add bootstrapping on top of reweighting and filtering (ablation) | BootGen · UTR | average score: 0.621 -> 0.698 | Improved |
| Use diverse aggregation of multiple generators (ablation) | BootGen · UTR | average score: 0.652 -> 0.698 | Improved |
| Use diverse aggregation of multiple generators (ablation) | BootGen · UTR | diversity: 17.83 -> 31.57 | Improved |
| Use diverse aggregation of multiple generators (ablation) | BootGen · UTR | score std of average score: 0.081 -> 0.001 | Improved |
| Compare BootGen with GFN-AL | BootGen · UTR | average score: 0.580 -> 0.698 | Improved |
| Compare BootGen with GFN-AL | BootGen · GFP | 100th percentile score: 0.057 -> 0.865 | Improved |
Closest papers by meaning
- Improved Off-policy Reinforcement Learning in Biological Sequence Design (ICML 2025 · similarity 0.86)
- Designing Biological Sequences without Prior Knowledge Using Evolutionary Reinforcement Learning (AAAI 2024 · similarity 0.85)
- Absorb & Escape: Overcoming Single Model Limitations in Generating Heterogeneous Genomic Sequences (NeurIPS 2024 · similarity 0.85)
- Curriculum Learning for Biological Sequence Prediction: The Case of De Novo Peptide Sequencing (ICML 2025 · similarity 0.84)
- Steering Generative Models with Experimental Data for Protein Fitness Optimization (NeurIPS 2025 · similarity 0.84)
- Bio-xLSTM: Generative modeling, representation and in-context learning of biological and chemical sequences (ICLR 2025 · 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.
