Research map / Architectures and efficiency / RNNs and normalization
Adaptive Hopfield Network: Rethinking Similarities in Associative Memory
Shurong Wang, Yuqi Pan, Zhuoyang Shen, Meng Zhang, Hongwei Wang, Guoqi Li
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 similarity with learned adaptive similarity | Hopfield network (M-Hop) · Mixed-variant retrieval (synthetic, difficulty 0.4) | retrieval accuracy: .520 -> .724 | Improved |
| replace fixed similarity with learned adaptive similarity | Hopfield network (M-Hop) · Mixed-variant retrieval (MNIST, difficulty 0.7) | retrieval accuracy: .661 -> .849 | Improved |
| use adaptive similarity instead of kernel optimized with the same loss (K2-Hop) | Hopfield network · Heart | accuracy: .6473 -> .7315 | Improved |
| replace M-Hop with A-Hop as a classifier | Hopfield network · Adult | accuracy: .8080 -> .8634 | Improved |
| A-Hop versus gradient boosting | A-Hop · Tabular classification (5 datasets) | accuracy: Adult .8634 vs XGBoost .8640 | Mixed |
| use A-Hop as HopfieldLayer in image classifier | K-Hop · CIFAR100 | accuracy: .2877 -> .2904 | Improved |
| use A-Hop as HopfieldPooling in MIL | S-Hop · MIL Fox | AUC-ROC: .6433 -> .6753 | Improved |
Closest papers by meaning
- Uniform Memory Retrieval with Larger Capacity for Modern Hopfield Models (ICML 2024 · similarity 0.89)
- In-Context Learning as Conditioned Associative Memory Retrieval (ICML 2025 · similarity 0.88)
- On Computational Limits of Modern Hopfield Models: A Fine-Grained Complexity Analysis (ICML 2024 · similarity 0.88)
- On the Recall Scaling Laws in Mamba: A Theoretical and Mechanistic Study via Hashing (NeurIPS 2026 · similarity 0.87)
- In-Context Denoising with One-Layer Transformers: Connections between Attention and Associative Memory Retrieval (ICML 2025 · similarity 0.87)
- HippoRAG: Neurobiologically Inspired Long-Term Memory for Large Language Models (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.
