AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

Efficient Test-time Inference for Generative Planning Models with OCL Search

Robert Gieselmann, Mihai Samson, Federico Pecora, Jeremy L Wyatt

ICML 2026 · Probabilistic and scientific modeling › Molecules and proteins · approach: Prediction-powered and conformal inference · 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
use OCLGen instead of MCTS with full rolloutsGPT-2 planner · Blocksworldmean plan length: 54.56 -> 43.88Improved
use OCLGen instead of Best-of-NGPT-2 planner · Labyrinthmean plan length: 17.42 -> 12.99Improved
add depth-partitioned selection (ablation)OCLGen · Blocksworldmean plan length: 48.81 -> 43.88Improved
add depth-partitioned selection (ablation)OCLGen · Sokobaninstances solved optimally: 353 / 489 (without) vs 377 / 489 (with); length 129.39 vs 128.67Mixed
add confidence-based adaptive expansion (ablation)OCLGen · Blocksworldmean plan length: 44.35 -> 43.88Improved
add confidence-based adaptive expansion (ablation)OCLGen · Sokobanmean plan length: 129.73 -> 128.67No clear effect
use lower-percentile cost-to-go instead of mode (ablation)OCLGen · Blocksworldmean plan length: 44.68 -> 43.88Improved
use lower-percentile cost-to-go instead of mode (ablation)heuristic model · Blocksworld heuristic MAEMAE: 8.68 -> 4.07Improved
scan vs uniform depth selection (ablation)OCLGen · 4 planning domainsmean plan length: 43.88 vs 44.10 (Blocksworld)No clear effect
generate self-improvement training plans with OCLGen instead of MCTSGPT-2 planner · Blocksworld, Best-of-N after 3 roundsmean plan length: 53.15 -> 41.40Improved
3 rounds of self-improvement with OCLGen searchOCLGen test-time search · Blocksworldinstances solved optimally: 528 / 630 -> 630 / 630Improved
self-improvement fine-tuning on search plans, Best-of-N completion on SokobanGPT-2 planner · Sokobancompletion rate: 95.8 -> 93.4Hurt

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