Research map / Architectures and efficiency / Attention and transformers
Energy Landscape-Aware Vision Transformers: Layerwise Dynamics and Adaptive Task-Specific Training via Hopfield States
Runze Xia, Richard Jiang
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Analyzes ViT layer dynamics through the lens of modern Hopfield networks and proposes a layer instability metric to guide adaptive, task-specific training.
Problem
Vision Transformers have deep, uniform layer stacks that incur significant computational overhead, and their layerwise dynamics are poorly understood.
Approach
Connects self-attention to modern Hopfield networks to interpret ViT layers as energy-based memory dynamics, and introduces a layer instability metric to characterize per-layer behavior for adaptive task-specific training.
Experimental setup
Builds on: modern Hopfield networks
Closest papers by meaning
- On the Role of Hidden States of Modern Hopfield Network in Transformer (NeurIPS 2025 · similarity 0.88)
- Bi-ViT: Pushing the Limit of Vision Transformer Quantization (AAAI 2024 · similarity 0.87)
- What Does It Mean to Be a Transformer? Insights from a Theoretical Hessian Analysis (ICLR 2025 · similarity 0.87)
- An Evolved Universal Transformer Memory (ICLR 2025 · similarity 0.86)
- Task-Free Dynamic Sparse Vision Transformer for Continual Learning (AAAI 2024 · similarity 0.86)
- On the Role of Attention Masks and LayerNorm in Transformers (NeurIPS 2024 · 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.
