MIT-States
PulseAugur coverage of MIT-States — every cluster mentioning MIT-States across labs, papers, and developer communities, ranked by signal.
2 day(s) with sentiment data
-
Vision models show "fragmented linearity" in understanding abstract concepts
Researchers have investigated whether vision models can grasp abstract concepts, similar to how language models understand abstract directions in embedding space. Their study of 46 vision encoders revealed that while mo…
-
New framework formalizes compositional reasoning in vision-language models
Researchers have introduced a new framework called Geodesically Decomposable Embeddings (GDE) that formalizes how large vision-language models encode compositional semantic structures. This framework identifies directio…
-
New CLEAR framework enhances Compositional Zero-Shot Learning
Researchers have introduced CLEAR, a novel framework for Compositional Zero-Shot Learning (CZSL) that addresses limitations in existing methods. CLEAR re-frames primitive variations as context-driven activations of visu…