AI Paper Atlas

Research map / Vision-language and multimodal / Robot action and navigation

Programmatically Grounded, Compositionally Generalizable Robotic Manipulation

Renhao Wang, Jiayuan Mao, Joy Hsu, Hang Zhao, Jiajun Wu, Yang Gao

ICLR 2023 · Vision-language and multimodal › Robot action and navigation · approach: Multimodal prompt-driven manipulation · 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 program-based modular grounding and action modules instead of CLIPort's monolithic language-conditioned policyrobotic manipulation · packing-color-box, 100 demonstrationssuccess score: 23.0 -> 72.0Improved
use program-based modular grounding and action modules instead of CLIPort's monolithic language-conditioned policyrobotic manipulation · packing-location-box, 100 demonstrationssuccess score: 11.0 -> 62.0Improved
use program-based modular grounding and action modules instead of CLIPort's monolithic language-conditioned policyrobotic manipulation · separating-location-piles, 100 demonstrationssuccess score: 48.1 -> 69.2Improved
use program-based modular grounding and action modules instead of CLIPort's monolithic language-conditioned policyrobotic manipulation · pushing-shapes, 100 demonstrationssuccess score: 2.7 -> 36.6Improved
use program-based modular grounding and action modules instead of CLIPort's monolithic language-conditioned policyrobotic manipulation · packing-prepositionssuccess score: 28.0 -> 45.0Improved
use program-based modular grounding and action modules instead of CLIPort's monolithic language-conditioned policyrobotic manipulation · packing-nested-prepositionssuccess score: 17.0 -> 31.0Improved
replace MaskCLIP grounding maps with ground-truth segmentation maps (ablation)ProgramPort (single) · packing-shapes, 100 demonstrationssuccess score: 52.0 -> 97.0Improved
replace MaskCLIP grounding maps with ground-truth segmentation maps (ablation)ProgramPort (single) · packing-google-objects-seq, 100 demonstrationssuccess score: 78.4 -> 95.8Improved
replace MaskCLIP grounding maps with ground-truth segmentation maps (ablation)ProgramPort (single) · assembling-kits, 100 demonstrationssuccess score: 40.6 -> 92.0Improved
replace MaskCLIP grounding maps with ground-truth segmentation maps (ablation)ProgramPort (single) · separating-small-piles, 100 demonstrationssuccess score: 76.8 -> 80.6Improved
replace MaskCLIP grounding maps with ground-truth segmentation maps (ablation)ProgramPort (single) · put-blocks-in-bowls, 100 demonstrationssuccess score: 38.2 -> 97.2Improved
train on 100 rather than 1000 demonstrations for packing-shapesProgramPort versus CLIPort · packing-shapessuccess score: +25.0 pointsImproved

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