The paper proposes that composing typed neuro-symbolic concepts (programs plus neural embeddings) yields data-efficient, compositionally generalizable, continually learnable, and zero-shot transferable AI agents.
Tenenbaum, and Chuang Gan
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Neuro-Symbolic Concepts
The paper proposes that composing typed neuro-symbolic concepts (programs plus neural embeddings) yields data-efficient, compositionally generalizable, continually learnable, and zero-shot transferable AI agents.