AI Paper Atlas

Research map / Probabilistic and scientific modeling / Variational and generative sampling

Generative Modeling of Discrete Latent Structures via Dynamic Policy Gradients

Stefan Ivanovic, Ge Liu, Mohammed El-Kebir

ICML 2026 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Disentangled and discrete latent models

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 dynamic reward rescaling by 1/Pr(X_i|theta) (ablation)naive policy gradient · Process graph inferenceF1: 0 -> 0.891 (k=10)Improved
add dynamic reward rescaling by 1/Pr(X_i|theta) (ablation)naive policy gradient · Subset inference (|U|=1000)median F1: 0.849 -> 0.938Improved
replace local search with GReinSS (shared latent model)local search · Subset inference (|U|=1000)median F1: 0.869 -> 0.938Improved
use GReinSS instead of GEM-based VAE, autoregressive, diffusionGEM baselines · Subset inference (|U|=1000)median F1: < 0.4 -> 0.938Improved
use GReinSS instead of baselines with very few random walksbaselines · Process graph inference (k=10)median F1: below 0.55 -> 0.891Improved
use GReinSS instead of GFlowNetsGFlowNets · Process graph inferencemedian F1: below 0.55 -> 0.891Improved
increase number of random walks kmost methods · Process graph inferenceF1Improved
use a different off-policy sampling proposal biased toward Pr(S|X,theta) (ablation)GReinSS · Subset inferenceF1No clear effect
use GReinSS instead of RSEMRSEM · RNA isoform reconstruction (MBD2, 61 samples)isoform prediction error: 0.471 -> 0.0069Improved
use GReinSS instead of RSEM across genesRSEM · RNA isoform reconstruction (14,390 genes)isoform prediction error: median difference -0.0405Improved
use GReinSS instead of GEM baselines at tiny universe sizeGEM baselines · Subset inference (|U|=10)median F1: 1.0No clear effect

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