CLMN: Concept based Language Models via Neural Symbolic Reasoning

ArXi:2510.10063v2 Announce Type: replace-cross Deep learning has advanced NLP, but interpretability remains limited, especially in healthcare and finance. Concept bottleneck models tie predictions to human concepts in vision, but NLP versions either use binary activations that harm text representations or latent concepts that weaken semantics, and they rarely model dynamic concept interactions such as negation and context. We