Weight Tying Biases Token Embeddings Towards the Output Space
By Antonio Lopardo, Avyukth Harish, Catherine Arnett, Akshat Gupta. 2026
Subfields: Large Language Models, Representation & Self-Supervised Learning, Model Architectures
By Antonio Lopardo, Avyukth Harish, Catherine Arnett, Akshat Gupta. 2026
Subfields: Large Language Models, Representation & Self-Supervised Learning, Model Architectures