Research map / Architectures and efficiency / RNNs and normalization
Coupling Liquid Time-Constant Encoders with Modern Hopfield Memory
Bishal Ranjan Swain, Kyung Joo Cheoi, Jaepil Ko
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 |
|---|---|---|---|
| couple Hopfield memory retrieval to LTC (ablation) | LTC · ablation avg over datasets | accuracy: 87.99 -> 90.42 | Improved |
| match LTC parameter count without memory (ablation) | LTC · ablation avg over datasets | accuracy: 87.99 -> 88.47 | No clear effect |
| freeze Hopfield weights at initialization (ablation) | LTC-MHN · ablation avg over datasets | accuracy: 90.42 -> 88.03 | Hurt |
| set beta to 0 (uniform retrieval) (ablation) | LTC-MHN · ablation avg over datasets | accuracy: 90.42 -> 82.21 | Hurt |
| shuffle memory indices each epoch (ablation) | LTC-MHN · ablation avg over datasets | accuracy: 90.42 -> 80.26 | Hurt |
| add Hopfield memory to LTC (ablation) | LTC · ablation avg over datasets | Hessian trace: 1.6x10^-2 -> 1.55x10^-3 | Improved |
| add Hopfield memory to LTC | LTC · Gesture | accuracy: 68.45% -> 71.23% | Improved |
| add Hopfield memory to LTC | LTC · Sequential MNIST | accuracy: 95.36% -> 97.30% | Improved |
| add Hopfield memory to LTC | LTC · Ozone | F1-score: 0.304 -> 0.321 | Improved |
| add Hopfield memory to LTC | LTC · Power | squared-error: 0.592 -> 0.629 | Hurt |
Closest papers by meaning
- Continual Learning in Modern Hopfield Networks with an Application to Diffusion Models (NeurIPS 2026 · similarity 0.86)
- Uniform Memory Retrieval with Larger Capacity for Modern Hopfield Models (ICML 2024 · similarity 0.85)
- Relating Hopfield Networks to Episodic Control (NeurIPS 2024 · similarity 0.85)
- Exponential Dynamic Energy Network for High Capacity Sequence Memory (NeurIPS 2025 · similarity 0.85)
- STanHop: Sparse Tandem Hopfield Model for Memory-Enhanced Time Series Prediction (ICLR 2024 · similarity 0.85)
- In-Context Denoising with One-Layer Transformers: Connections between Attention and Associative Memory Retrieval (ICML 2025 · similarity 0.85)
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.
