Publications

3 results for Taku Ito

Learning interpretable positional encodings in transformers depends on initialization
- - Taku Ito
  - Luca Cocchi
  - et al.
- 2025
- ICML 2025
Geometry of naturalistic object representations in recurrent neural network models of working memory
- - Xiaoxuan Lei
  - Taku Ito
  - et al.
- 2024
- NeurIPS 2024
On the generalization capacity of neural networks during generic multimodal reasoning
- - Taku Ito
  - Soham Dan
  - et al.
- 2024
- ICLR 2024