Research map / Graph learning / Knowledge and temporal graphs
FEAST: Fully Connected Expressive Attention for Spatial Transcriptomics
Taejin Jeong, Joohyeok Kim, Jinyeong Kim, Chanyoung Kim, Seong Jae Hwang
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 |
|---|---|---|---|
| add negative-aware attention (ablation) | FEAST (no off-grid) · Her2ST | PCC: 0.5396 -> 0.5464 | Improved |
| add off-grid sampling (pseudo-spots) (ablation) | FEAST (no negative-aware) · Her2ST | PCC: 0.5396 -> 0.5458 | Improved |
| combine negative-aware attention with off-grid sampling (ablation) | FEAST · Her2ST | PCC: 0.5396 -> 0.5524 | Improved |
| increase local kNN k from 8 to 32 (ablation) | FEAST · Her2ST | PCC: 0.5453 -> 0.5524 | Improved |
| increase local kNN k from 32 to 64 or 100 (ablation) | FEAST · Her2ST | PCC: 0.5524 -> 0.5481 (k=64), 0.5484 (k=100) | Hurt |
| fully connected attention model vs MERGE sparse-graph GNN | FEAST · Her2ST | PCC: 0.5037 -> 0.5524 | Improved |
| FEAST vs HisToGene on SCC error metrics | FEAST · SCC | MSE: 0.3095 (HisToGene) vs 0.4501 (FEAST) | Hurt |
Closest papers by meaning
- Global Context-aware Representation Learning for Spatially Resolved Transcriptomics (ICML 2025 · similarity 0.89)
- MERGE: Multi-faceted Hierarchical Graph-based GNN for Gene Expression Prediction from Whole Slide Histopathology Images (CVPR 2025 · similarity 0.89)
- Scalable Generation of Spatial Transcriptomics from Histology Images via Whole-Slide Flow Matching (ICML 2025 · similarity 0.88)
- Multi-modal Topology-embedded Graph Learning for Spatially Resolved Genes Prediction from Pathology Images with Prior Gene Similarity Information (CVPR 2025 · similarity 0.88)
- Diffusion Generative Modeling for Spatially Resolved Gene Expression Inference from Histology Images (ICLR 2025 · similarity 0.88)
- Preserving DEG Rankings for Gene Discovery in Histology-Based Spatial Gene Expression Prediction (NeurIPS 2026 · similarity 0.88)
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.
