Molecular Property Optimization on ppi inhibitor
86Normalized Rewardπ(·|y*)
Evaluation Results
| Method | Links | |
|---|---|---|
| π(·|y*)2026.04 | 86 | |
| RLRL training budget=20002026.04 | 70 | |
| RLRL training budget=10002026.04 | 62 | |
| Reward Weighted Classifier-Free Guidance (RCFG)Inference-time sampling budget |YS|=642026.04 | 55 | |
| Reward Weighted Classifier-Free Guidance (RCFG)Inference-time sampling budget |YS|=322026.04 | 49 | |
| RLRL training budget=5002026.04 | 49 | |
| Reward Weighted Classifier-Free Guidance (RCFG)Inference-time sampling budget |YS|=162026.04 | 46 | |
| Reward Weighted Classifier-Free Guidance (RCFG)Inference-time sampling budget |YS|=82026.04 | 36 | |
| Reward Weighted Classifier-Free Guidance (RCFG)Inference-time sampling budget |YS|=42026.04 | 23 | |
| Reward Weighted Classifier-Free Guidance (RCFG)Inference-time sampling budget |YS|=22026.04 | 13 |