One line of work does so by studying their ability to distinguish non-isomorphic graphs.

Latent embeddings offer an effective way of defining a content gap.

We introduce Lifelong ICL, a problem setting that challenges long-context language models (LMs) to learn a sequence of language tasks through in-context learning(ICL).

We also collect official crash reports from the Tennessee DepartmentofTransportation TrafficManagement Centerandmanuallylabelall other potential anomalies inthe dataset.