Research map / Architectures and efficiency / RNNs and normalization
Recurrent Equivariant Constraint Modulation: Learning Per-Layer Symmetry Relaxation from Data
Stefanos Pertigkiozoglou, Mircea Petrache, Shubhendu Trivedi, Kostas Daniilidis
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 |
|---|---|---|---|
| replace fixed-schedule relaxation (ES) with RECM | VN-PointNet · ModelNet40 Rotated | instance accuracy: 0.72 -> 0.80 | Improved |
| replace fixed-schedule relaxation (ES) with RECM | VN-PointNet · ModelNet40 Aligned | instance accuracy: 0.67 -> 0.90 | Improved |
| replace Bayesian relaxation (ELLA) with RECM | VN-PointNet · ModelNet40 Rotated | instance accuracy: 0.78 -> 0.80 | Improved |
| replace Bayesian relaxation (ELLA) with RECM | VN-PointNet · ModelNet40 Aligned | instance accuracy: 0.90 -> 0.90 | No clear effect |
| add RECM relaxation on top of SEGNN | SEGNN · N-body simulation | MSE: 5.6 -> 3.7 | Improved |
| replace ACE-exact/appr with RECM | SEGNN · N-body simulation | MSE: 3.8 -> 3.7 | No clear effect |
| add RECM on top of EGNO | EGNO · Motion capture (Run) | MSE: 33.9 -> 22.6 | Improved |
| add RECM on top of EGNO | EGNO · Motion capture (Walk) | MSE: 8.1 -> 6.6 | Improved |
| add RECM to equivariant ETFlow | ETFlow · GEOM-Drugs | mean precision: 74.38 -> 75.10 | Improved |
| add RECM to equivariant ETFlow | ETFlow · GEOM-Drugs | mean recall: 79.53 -> 79.44 | No clear effect |
| add RECM to equivariant DiTMC-Eq | DiTMC · GEOM-Drugs | mean precision: 75.3 -> 76.1 | Improved |
| use RECM instead of fully unconstrained DiTMC-Unc | DiTMC · GEOM-Drugs | mean precision: 76.5 -> 76.1 | Hurt |
Closest papers by meaning
- Improving Equivariant Model Training via Constraint Relaxation (NeurIPS 2024 · similarity 0.92)
- Learning (Approximately) Equivariant Networks via Constrained Optimization (NeurIPS 2025 · similarity 0.91)
- A Tale of Two Symmetries: Exploring the Loss Landscape of Equivariant Models (NeurIPS 2025 · similarity 0.90)
- Improving Equivariant Networks with Probabilistic Symmetry Breaking (ICLR 2025 · similarity 0.88)
- Flow Equivariant Recurrent Neural Networks (NeurIPS 2025 · similarity 0.88)
- Approximately Equivariant Neural Processes (NeurIPS 2024 · similarity 0.87)
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.
