Research map / Architectures and efficiency / RNNs and normalization
Learning to Remember More with Less Memorization
Hung Le, Truyen Tran, Svetha Venkatesh
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 regular writes with uniform writing (ablation) | DNC · Copy L=50 | accuracy: 68.0 -> 97.7 | Improved |
| replace regular writes with uniform writing (ablation) | DNC · Reverse L=50 | accuracy: 65.0 -> 100 | Improved |
| use cached local attention with uniform writing rather than uniform writing alone (ablation) | DNC+UW · Copy L=50 | accuracy: 97.7 -> 83.8 | Hurt |
| use cached local attention with uniform writing rather than uniform writing alone (ablation) | DNC+UW · Add L=50 | accuracy: 84.8 -> 94.4 | Improved |
| use cached local attention with uniform writing rather than uniform writing alone (ablation) | DNC+UW · Max L=50 | accuracy: 71.7 -> 82.3 | Improved |
| apply uniform writing rather than regular writing (ablation) | DNC · pMNIST | accuracy: 94.0 -> 95.6 | Improved |
| apply cached uniform writing rather than uniform writing (ablation) | DNC · MNIST | accuracy: 98.6 -> 99.1 | Improved |
| apply cached uniform writing rather than uniform writing (ablation) | DNC · IMDb | accuracy: 91.4 -> 91.3 | Hurt |
| apply cached uniform writing rather than uniform writing (ablation) | DNC · Yelp Review Polarity | accuracy: 96.4 -> 96.4 | No clear effect |
| use uniform writing instead of regular writing (ablation) | DNC · Double task | training time reduction: up to 60% for DNC and 28% for NTM | Improved |
Closest papers by meaning
- Memorization Sinks: Isolating Memorization during LLM Training (ICML 2025 · similarity 0.88)
- On the Optimal Memorization Power of ReLU Neural Networks (ICLR 2022 · similarity 0.88)
- Predictive Attractor Models (NeurIPS 2024 · similarity 0.88)
- Do Your Best and Get Enough Rest for Continual Learning (CVPR 2025 · similarity 0.87)
- Blending Complementary Memory Systems in Hybrid Quadratic-Linear Transformers (NeurIPS 2025 · similarity 0.87)
- Unlocking the Power of Rehearsal in Continual Learning: A Theoretical Perspective (ICML 2025 · similarity 0.86)
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.
