Chengfeng Wu
2 indexed papers
Publications per year
Top categories
Frequent co-authors
Research Timeline
CORE-MTL proposes a representation-centric framework that uses causal orthogonal representations to disentangle task-relevant structure from nuisance variation in multi-task learning, achieving superior generalization.
CORE-MTL proposes a representation-centric framework that uses causal orthogonal representations to disentangle task-relevant structure from nuisance variation in multi-task learning, achieving superior generalization.
Papers
CORE-MTL: Rethinking Gradient Balancing via Causal Orthogonal Representations
CORE-MTL proposes a representation-centric framework that uses causal orthogonal representations to disentangle task-relevant structure from nuisance variation in multi-task learning, achieving superi…