Where to Bind Matters: Hebbian Fast Weights in Vision Transformers for Few-Shot Character Recognition
For researchers in few-shot learning and meta-learning, this work provides an empirical analysis of Hebbian plasticity in transformer architectures, but the gains are incremental.
The paper studies Hebbian Fast-Weight modules in Vision Transformers for few-shot character recognition on Omniglot, finding that a single module placed after the final stage of Swin-Tiny achieves the highest accuracy (96.2% 1-shot, 99.2% 5-shot), outperforming its non-Hebbian baseline by 0.3 percentage points at 1-shot.
Standard transformer architectures learn fixed slow-weight representations during training and lack mechanisms for rapid adaptation within an episode. In contrast, biological neural systems address this through fast synaptic updates that form transient associative memories during inference, a property known as Hebbian plasticity. In this paper, we conduct an empirical study of Hebbian Fast-Weight (HFW) modules integrated into multiple transformer backbones, including ViT-Small, DeiT-Small, and Swin-Tiny. We evaluate six model variants: ViT, DeiT, Swin, ViT-Hebbian, DeiT-Hebbian, and Swin-Hebbian on 5-way 1-shot and 5-way 5-shot classification tasks using the Omniglot benchmark under a Prototypical Network meta-learning framework. We propose a single module placement strategy for Swin-Tiny in which one HFW module is applied to the final stage feature map after all hierarchical stages have completed. This design avoids the training instability caused by placing separate Hebbian modules at each stage and achieves the highest test accuracy across all six models (96.2\% at 1-shot; 99.2\% at 5-shot), outperforming its non-Hebbian baseline by $+0.3$ percentage points at 1-shot. We analyze the interaction between Swin's shifted window inductive bias and episode-level Hebbian binding, discuss why per-block placement fails for ViT and DeiT variants in a low-data regime, and situate the results within the wider literature on fast and slow-weight meta-learning.