AI Paper Atlas

Research map / Graph learning / Graph self-supervised learning

SIR: Structured Image Representations for Explainable Robot Learning

Paul Mattes, Jan Schwab, Jens Bosch, Maximilian Xiling Li, Nils Blank, Minh-Trung Tang, Moritz Haberland, Rudolf Lioutikov

CVPR 2026 · Graph learning › Graph self-supervised learning · approach: Multi-view graph clustering · 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 fully-connected scene graph instead of image embeddingMDT · RoboCasa 24 taskssuccess rate: 14.81 -> 16.98Improved
add learned instruction-grounded sparsification to FC graph (ablation)MDT graph policy · RoboCasa 24 taskssuccess rate: 16.98 -> 19.50Improved
graph representation for Drawers and Pick/Place tasksMDT · RoboCasa Drawerssuccess rate: 49.75 -> 46.25Hurt
graph representation for LeversMDT · RoboCasa Leverssuccess rate: 23.67 -> 48.50Improved
add soft histogram loss to node removal (ablation)SIR sparsification · RoboCasa 24 taskssuccess rate: 9.60 -> 19.50Improved
random node removal instead of learned (ablation)SIR sparsification · RoboCasa 24 taskssuccess rate: 19.50 -> 5.48Hurt
use task-specific k and L1 node-weight guidance (SIR vs TopK) (ablation)SIR sparsification · RoboCasa 24 taskssuccess rate: 18.44 -> 19.50Improved
use threshold sparsification instead of FC (ablation)FC graph · RoboCasa 24 taskssuccess rate: 16.98 -> 17.17No clear effect
cropped image features as node features vs label+bbox (ablation)Split-View graph · RoboCasa 24 taskssuccess rate: 10.98 -> 16.65Improved
use point clouds via GNN nodes instead of directly into action generator (ablation)MDT · RoboCasa 24 taskssuccess rate: 4.13 -> 11.08Improved
introduce 3-9 distractor objects at inference (ablation)Image baseline · RoboCasa with distractorssuccess rate: drop of 3.3% on averageHurt
introduce distractor objects at inference (ablation)SIR / FC-Graph · RoboCasa with distractorssuccess rate: almost no degradationNo 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