← Search

Walter Nelson

2 accepted papers

2026

Identifiability and recoverability in self-supervised models

ICLR 2026poster

Self-supervised models exhibit a surprising stability in their internal representations. Whereas most prior work treats this stability as a single property, we formalize it as two distinct concepts: **statistical identifiability** (consistency of representations across runs) and **structural identi…

Cited by 0SourcecodeScholar