Research map / Graph learning / Graph self-supervised learning
Learning Graph-Structured Sum-Product Networks for Probabilistic Semantic Maps
Kaiyu Zheng, Andrzej Pronobis, Rajesh Rao
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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use GraphSPN instead of MRF-2 | Semantic class inference for placeholders · Freiburg, noise levels 2 and 5 | accuracy: 67.58%(+/-10.42) vs 28.32%(+/-7.53); 40.59%(+/-12.22) vs 24.23%(+/-11.40) | Improved |
| Use GraphSPN instead of MRF-2 | Semantic class inference for placeholders · Saarbrücken, noise levels 2 and 5 | accuracy: 78.15%(+/-9.95) vs 39.85%(+/-19.42); 55.18%(+/-19.67) vs 30.58%(+/-5.57) | Improved |
| Use GraphSPN instead of MRF-2 | Semantic class inference for placeholders · Stockholm, noise levels 2 and 5 | accuracy: 67.57%(+/-11.11) vs 12.44%(+/-3.46); 37.56%(+/-10.44) vs 10.04%(+/-2.59) | Improved |
| Use three-variable rather than pairwise potentials | MRF semantic place inference · Semantic place classification with noisy local evidence | accuracy | Hurt |
| Use GraphSPN rather than MRF models | Novel spatial configuration detection · Semantic maps with certain label assignments | ROC performance | Mixed |
Closest papers by meaning
- Sum-Product-Set Networks: Deep Tractable Models for Tree-Structured Graphs (ICLR 2024 · similarity 0.87)
- Beyond Coordinates: Encoding Graph Structure via Contextual Distribution and Relational Similarity (NeurIPS 2026 · similarity 0.86)
- Tractable Probabilistic Graph Representation Learning with Graph-Induced Sum-Product Networks (ICLR 2024 · similarity 0.86)
- Semantic Routing via Autoregressive Modeling (NeurIPS 2024 · similarity 0.85)
- Random-Set Graph Neural Networks (NeurIPS 2026 · similarity 0.85)
- The Underappreciated Power of Vision Models for Graph Structural Understanding (NeurIPS 2025 · similarity 0.84)
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.
