Search papers, labs, and topics across Lattice.
King Abdullah University of Science and Technology (KAUST)
2
0
2
Video foundation models don't need larger web scrapes to master fine-grained spatiotemporal understanding—they need multi-granularity synthetic captions paired with view-specific [CLS] embeddings.
Machine translation benchmarks have functionally saturated, but pairing human-authored failure cases with deterministic verification rules reveals critical multimodal blind spots that automated metrics consistently miss.
