Research map / Large language models / Agents, code and math
Steering Large Language Models through the DMTA Cycle: Structure-Based Drug Design via Knowledge-Driven Bi-Level Thompson Sampling
Xuanning Hu, Hao Tuo, Jinglong Ji, Anchen Li, Qianli Xing, Bo Yang
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 upper-level Bayesian seed selection (ablation) | K-BTS-rand · CrossDocked2020 (first 10 targets, K-BTS-rand ablation) | Top-1 avg Vina docking score: -8.95 -> -9.83 | Improved |
| add upper-level Bayesian seed selection (ablation) | K-BTS-rand · CrossDocked2020 (first 10 targets, K-BTS-rand ablation) | Top-20 avg Vina docking score: -7.79 -> -9.14 | Improved |
| add lower-level Thompson sampling of actions (ablation) | K-BTS-rand · CrossDocked2020 (first 10 targets, K-BTS-rand ablation) | Top-1 avg Vina docking score: -9.61 -> -9.83 | Improved |
| add knowledge base of chemical actions (ablation) | K-BTS-rand · CrossDocked2020 (first 10 targets, K-BTS-rand ablation) | Top-1 avg Vina docking score: -9.54 -> -9.83 | Improved |
| add warm-start of knowledge base from initial molecules (ablation) | K-BTS-rand · CrossDocked2020 (first 10 targets, K-BTS-rand ablation) | Top-1 avg Vina docking score: -9.74 -> -9.83 | Improved |
| retrieved knowledge actions applied stochastically instead of free LLM exploration (ablation) | K-BTS-rand · CrossDocked2020 (first 10 targets, K-BTS-rand ablation) | Top-1 avg Vina docking score: -9.54 -> -9.61 | Improved |
| full K-BTS convergence reduces diversity (ablation) | K-BTS-rand · CrossDocked2020 (first 10 targets, K-BTS-rand ablation) | internal diversity avg: 0.78 -> 0.50 | Hurt |
| replace ELILLM with K-BTS iterative refinement from diffusion starting points | K-BTS-diff · CrossDocked2020 test (100 targets) | Top-1 avg Vina docking score: -9.80 -> -10.23 | Improved |
| start from random training-set ligands instead of diffusion-generated molecules | K-BTS · CrossDocked2020 test (100 targets) | Top-1 avg Vina docking score: -10.23 -> -9.60 | Hurt |
| use K-BTS retrieved action heuristics vs ELILLM | K-BTS-rand · CrossDocked2020 test (100 targets) | SA avg: 0.63 -> 0.75 | Improved |
Closest papers by meaning
- Generating Novel Leads for Drug Discovery Using LLMs with Logical Feedback (AAAI 2024 · similarity 0.86)
- Enhancing Ligand Validity and Affinity in Structure-Based Drug Design with Multi-Reward Optimization (ICML 2025 · similarity 0.85)
- Efficient Evolutionary Search Over Chemical Space with Large Language Models (ICLR 2025 · similarity 0.85)
- TurboHopp: Accelerated Molecule Scaffold Hopping with Consistency Models (NeurIPS 2024 · similarity 0.84)
- Drug-TTA: Test-Time Adaptation for Drug Virtual Screening via Multi-task Meta-Auxiliary Learning (ICML 2025 · similarity 0.84)
- Controllable 3D Molecular Generation for Structure-Based Drug Design Through Bayesian Flow Networks and Gradient Integration (NeurIPS 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.
