<ul data-eligibleForWebStory="false"><li>Multi-Task Learning (MTL) in shared networks can lead to negative transfer due to differences in task objectives.</li><li>Pre-trained transformers have limitations in adaptability, motivating the development of Dynamic Token Modulation and Expansion (DTME-MTL).</li><li>DTME-MTL addresses gradient conflicts in token space to enhance adaptability and reduce overfitting without duplicating network parameters.</li><li>Experiments show that DTME-MTL offers a scalable and efficient solution for improving transformer-based MTL models.</li></ul>

Resolving Token-Space Gradient Conflicts: Token Space Manipulation for Transformer-Based Multi-Task Learning

Discover more