AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

Constrained Molecule Generation Modelled Using the Grammar Constraint

David Saikali, Gilles Pesant

AAAI 2026 · Probabilistic and scientific modeling › Molecules and proteins · approach: Structure-based 3D molecule generation · 1 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 belief-propagation marginals (CPBP) instead of domWDeg searchCP model of SMILES generation · 81 constrained molecule instancesinstances solved: 52 -> 72Improved
use belief-propagation marginals (CPBP) instead of domWDeg searchCP model of SMILES generation · 81 constrained molecule instancesaverage failed nodes: 333.2 -> 8.1Improved
use belief-propagation marginals (CPBP) instead of domWDeg searchCP model of SMILES generation · 81 constrained molecule instancesaverage time (s): 231.2 -> 175.3Improved
add the weighted counting algorithm for the CFG constraint to CPBP (ablation)CPBP · 81 constrained molecule instancesinstances solved: 49 -> 72Improved
add the weighted counting algorithm for the CFG constraint to CPBP (ablation)CPBP · 81 constrained molecule instancesbacktrack-free generations: 3 -> 25Improved
add the weighted counting algorithm for the CFG constraint to CPBP (ablation)CPBP · 81 constrained molecule instancesaverage failed nodes: 9.9 -> 8.1Improved
add the weighted counting algorithm for the CFG constraint to CPBP (ablation)CPBP · 81 constrained molecule instancesaverage time (s): 298.2 -> 175.3Improved
use belief propagation guidance instead of classical CPCP model of SMILES generation · 81 constrained molecule instancesslope of time versus search effortMixed

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