Molecular Property Optimization on selective 3d druglike
0.81Normalized Rewardπ(·|y*)
Evaluation results
| Method | Links | |
|---|---|---|
| π(·|y*)2026.04 | 0.81 | |
| RLRL training budget=20002026.04 | 0.8 | |
| RLRL training budget=10002026.04 | 0.79 | |
| RLRL training budget=5002026.04 | 0.69 | |
| Reward Weighted Classifier-Free Guidance (RCFG)Inference-time sampling budget |YS|=642026.04 | 0.68 | |
| Reward Weighted Classifier-Free Guidance (RCFG)Inference-time sampling budget |YS|=162026.04 | 0.67 | |
| Reward Weighted Classifier-Free Guidance (RCFG)Inference-time sampling budget |YS|=322026.04 | 0.67 | |
| Reward Weighted Classifier-Free Guidance (RCFG)Inference-time sampling budget |YS|=82026.04 | 0.59 | |
| Reward Weighted Classifier-Free Guidance (RCFG)Inference-time sampling budget |YS|=42026.04 | 0.46 | |
| Reward Weighted Classifier-Free Guidance (RCFG)Inference-time sampling budget |YS|=22026.04 | 0.24 |