🛰️ Daily AI Frontier
‹ back to 2026-08-29

Representation Measurements Under Function-Preserving Reparameterizations

Research Representation Analysis

Ranking

Overall 79
Content 95
Popularity 42

Observed public metrics from 1 member.

Representative image for Representation Measurements Under Function-Preserving Reparameterizations

Merged summary

TL;DR - This paper shows that column-permutation parallel analysis can produce representation measurements that change under function-preserving reparameterizations of language models. Consequently, its component counts may reflect arbitrary hidden-coordinate choices rather than intrinsic model properties.

  • Across five models, three retrieval domains, and 75 transformations, median disagreement was 0.79 for component counts and 0.26 for fixed-threshold decisions.
  • A centering-only control changed 1,141 of 1,200 component counts despite leaving the observed covariance spectrum unchanged.
  • Independent parallel-analysis seeds preserved all corresponding control decisions, isolating the instability to the data-internal reference procedure.
  • Orthogonally invariant comparator scores remained numerically stable while achieving similar held-out discrimination.

Sources (1)

Representation Measurements Under Function-Preserving Reparameterizations

arXiv stat.ML Abdullah Karasan 2026-08-27 arXiv:2608.27020
Public signals Semantic Scholar citations 0 · Semantic Scholar influential citations 0
Providers: Hugging Face · N/A OpenAlex · N/A Publisher · N/A Semantic Scholar · Citations 0 · Influential citations 0 X · N/A Fetched 2026-09-25 14:26:23.853462 UTC

TL;DR - This paper shows that column-permutation parallel analysis can produce representation measurements that change under function-preserving reparameterizations of language models. Consequently, its component counts may reflect arbitrary hidden-coordinate choices rather than intrinsic model properties.

  • Across five models, three retrieval domains, and 75 transformations, median disagreement was 0.79 for component counts and 0.26 for fixed-threshold decisions.
  • A centering-only control changed 1,141 of 1,200 component counts despite leaving the observed covariance spectrum unchanged.
  • Independent parallel-analysis seeds preserved all corresponding control decisions, isolating the instability to the data-internal reference procedure.
  • Orthogonally invariant comparator scores remained numerically stable while achieving similar held-out discrimination.
item →