Get Started
Research questionHow can AI systems build reusable representations for learning novel concepts from sparse data?AI systems often struggle to transfer knowledge from a handful of examples because their learned representations do not capture the structure needed for systematic generalization.
AI
Code Generation & Program Synthesis
Machine Learning
Latest papersRecent research connected to this question, newest first.Towards a universal language of concepts: A surveyThis survey reviews computational concept-learning models that represent concepts as programs and evaluates their contribution toward a universal representational language. Its evidence concerns surveyed models rather than a single deployment setting.research paper · Sep 3, 2026
Related questions
How can pretrained models learn new concepts from evolving streams without task identities or replay while retaining prior knowledge?How can in-context learning select demonstrations by flexible reasoning logic rather than surface similarity?How can video models recognize unseen actions from only a few labeled examples?How can low-level vision generalist models adapt to previously unseen tasks from a single example?
Home
Topics
Search
Library