g.
Anchor distributions define context-specific similarity in contrastive models
By linking contrastive learning to information geometry, the authors show that a probability distribution over text anchors induces a Bregman divergence for tailored image retrieval.
Top University
Akash Kannan · Kiho Park · Victor Veitch
University of Chicago
Research Digest··3 min read
Kannan et al.
Why this paper
From University of Chicago
In one line
Contrastive representations naturally contain a family of geometries that can be specialized by a distribution over semantic anchors.
What we could check
- ·No code link found
- ·No weights link found
- ·No dataset link found
- ·No compute details found
- ·No stated limitations found
- ·No benchmark numbers found
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§