AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

Exploring Synthesizable Chemical Space with Iterative Pathway Refinements

Seul Lee, Karsten Kreis, Srimukh Prasad Veccham, Meng Liu, Danny Reidenbach, Saee Gopal Paliwal, Weili Nie, Arash Vahdat

ICLR 2026 · Probabilistic and scientific modeling › Molecules and proteins · approach: Structure-based 3D molecule generation · 3 citations (OpenAlex)

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 changedWhereMeasuredEffect
use bidirectional (BU+TD) iterative cycle instead of unidirectional sampling (ablation)ReaSyn · synthesizable reconstructionreconstruction rateImproved
add Edit Bridge refinement to the BU+TD cycle (ablation)ReaSyn · synthesizable reconstructionreconstruction rateImproved
use SMILES-based non-hierarchical pathway representation instead of fingerprint-based hierarchical one (ablation)bottom-up ReaSyn vs SynFormer · synthesizable reconstructionreconstruction rateImproved
train one model with random BU and TD directions instead of two separate models (ablation)ReaSyn · Enamine reconstructionreconstruction rate (%): 95.2 -> 95.0No clear effect
use Edit Bridge coupling instead of uniform coupling (ablation)Edit Flow · 10,000 training pairsaligned rate (%): 2.59 -> 70.56Improved
use Edit Bridge coupling instead of uniform coupling (ablation)Edit Flow · 10,000 training pairsedit operations per sequence: 142.87 -> 30.01Improved
use bidirectional single model vs BU-only model on ZINC250k (ablation)ReaSyn autoregressive model, BU inference · ZINC250k reconstructionreconstruction rate (%): 51.3 -> 38.8Hurt
use bidirectional single model vs TD-only model on Enamine (ablation)ReaSyn autoregressive model, TD inference · Enamine reconstructionreconstruction rate (%): 85.2 -> 82.5Hurt
add ReaSyn projection to Graph GAGraph GA · TDC oracles averageaverage AUC top-10: 0.633 -> 0.633No clear effect
use ReaSyn projection instead of SynNet in synthesis-constrained optimizationGraph GA · TDC oracles averageaverage AUC top-10: 0.545 -> 0.633Improved
use ReaSyn instead of SynFormer for synthesizable projectionsynthesizable hit expansion for JNK3 · JNK3 hit expansionanalog, improve and success rateImproved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy