SynCB: A Synergy Concept-Based Model with Dynamic Routing Between Concepts and Complementary Neural Branches
For interpretable AI researchers, SynCB offers a hybrid model that improves both accuracy and intervention responsiveness, though it is an incremental extension of existing hybrid approaches.
SynCB combines a concept-based branch with a complementary neural branch via a dynamic routing module, achieving up to 3.9% higher task accuracy than a full neural baseline and up to 6.43% better intervention performance than competitors across five datasets.
Concept-based (CB) models provide interpretability and support test-time human intervention, while standard neural networks (NN) offer strong task performance but little transparency. Prior work has explored hybrid formulations that integrate concepts and additional representations to improve accuracy, often at the cost of human interventions. We introduce the \emph{Synergy Concept-Based Model (SynCB)} framework, that combines a CB branch with a complementary neural branch, and a trainable routing module that dynamically selects which branch to use for each input. Unlike prior models, which fuse residual and concept-based predictions, SynCB keeps the two branches distinct and coordinates them through the routing module. Moreover, both branches are learned jointly, allowing information sharing between the complementary neural branch and CB branches through their common backbone. To improve responsiveness to interventions, we further introduce a test-time intervention policy and a corresponding loss. Across five datasets and CB benchmarks, SynCB consistently achieves higher task accuracy while remaining more responsive to human interventions, surpassing the full neural baseline by up to 3.9 percentage points and exceeding the strongest competitor in intervention performance by up to 6.43 percentage points.