Search papers, labs, and topics across Lattice.
This paper introduces UNified In-Context Operator Networks (UNICON), a foundation model designed to exhibit numerical intelligence by inferring and applying knowledge from numerical contexts across various scientific and social systems. UNICON demonstrates the ability to achieve specialist-level performance in disciplines not included in its training data, particularly when combined with language-model agents, which enhances its predictive capabilities. The findings indicate that increasing the diversity of the training corpus significantly improves the model's generalization to unseen disciplines, establishing UNICON as a pivotal advancement in the field of numerical intelligence.
UNICON achieves specialist-level performance in previously unseen disciplines without retraining, redefining the boundaries of numerical intelligence in AI.
Intelligence is commonly understood as the ability to acquire and apply knowledge, adapt to unfamiliar situations and solve new problems. Large language models exhibit this capacity by inferring task-relevant knowledge from textual context and applying it to new tasks. Yet intelligence need not be confined to language. For scientific and social systems, we need models that acquire and apply knowledge from numerical context-an ability we call numerical intelligence. Here we introduce UNified In-Context Operator Networks (UNICON), a foundation model that exhibits numerical intelligence across disciplines. Using graph-based examples from a system as context, UNICON infers the predictive relation shared across them and applies it to queries from the same system. Across scientific and social systems, including those from disciplines absent from training, the same model approaches specialist performance without retraining. Combining UNICON with language-model agents yields further gains, enabling it to surpass state-of-the-art specialists in a discipline unseen in training. We further show that training-corpus diversity improves generalization to unseen disciplines. Together, these results establish UNICON as a foundation model of numerical intelligence and position it as a building block for a broader ecosystem of artificial intelligence.
