AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

SynLaD: Latent Diffusion for Generating Synthesizable Molecules Conditioned on 3D Pharmacophore Profiles

Miruna Cretu, John Bradshaw, Patricia Suriana, Saeed Saremi, Omar Mahmood, Kirill Shmilovich, Kangway V. Chuang, Vishnu Sresht, Colin A Grambow

ICML 2026 · Probabilistic and scientific modeling › Molecules and proteins · approach: Structure-based molecule design with diffusion · 0 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
add joint 3D-decoder training with the synthesis head3D decoder latent diffusion · Unconditional generationAiZynthFinder success: 0.59 -> 0.72Improved
add joint 3D training to the synthesis decodersynthesis decoder · Unconditional generationAiZynthFinder success: 0.81 -> 0.87Improved
use beam search (b=5) instead of sampling (ablation)synthesis decoder · Autoencoder reconstructionsynthesis match rate (%): 43.4 -> 63.4Improved
use beam width 5 instead of 1 (ablation)synthesis decoder · Autoencoder reconstructionsynthesis match rate (%): 47.6 -> 63.4Improved
increase beam width from 5 to 10 (ablation)synthesis decoder · Autoencoder reconstructionsynthesis match rate (%): 63.4 -> 59.5Hurt
raise 3D loss weight lambda from 0.5 to 0.92 (ablation)autoencoder · Autoencoder loss weights3D molecule match rate (%): 82.1 -> 98.5Improved
raise 3D loss weight lambda from 0.92 to 0.98 (ablation)autoencoder · Autoencoder loss weightsbuilding block match rate (%): 82.5 -> 80.6Hurt
raise KL weight beta from 1e-5 to 1e-4 (ablation)autoencoder · Autoencoder loss weights3D molecule match rate (%): 98.5 -> 80.6Hurt
use latent dim 16 instead of 8 (ablation)autoencoder, beam search · Autoencoder reconstructionsynthesis match rate (%): 58.4 -> 63.4Improved
use 100 sampling steps instead of 50 (ablation)latent DiT · Unconditional generationvalidity: 0.85 -> 0.89Improved
add classifier-free guidance (w=4.0) (ablation)SynLaD synthesis outputs · Lit-PCBA hit diversificationhits (avg.): 17.9 -> 11.7Hurt
add classifier-free guidance (w=0.5) (ablation)SynLaD · Lit-PCBA hit diversificationvalidity: 0.49 -> 0.56Improved

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