Pretraining and alignment induce asymmetric geometric traces in transformer weights because alignment updates concentrate in read pathways due to activation covariance while write pathways inherit less structure from alignment losses.
arXiv preprint arXiv:1905.12213 , year=
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 3verdicts
UNVERDICTED 3roles
background 1polarities
background 1representative citing papers
ALU uses public data to suppress unlearning cost quadratically while characterizing distribution mismatch effects, enabling mass unlearning with maintained utility.
Information defined as maximum-caliber deviation derives IIT 3.0 cause-effect repertoires from constrained entropy maximization and equates to prediction error under CLT and LDT.
citing papers explorer
-
Where Pretraining writes and Alignment reads: the asymmetry of Transformer weight space
Pretraining and alignment induce asymmetric geometric traces in transformer weights because alignment updates concentrate in read pathways due to activation covariance while write pathways inherit less structure from alignment losses.
-
Unlearning with Asymmetric Sources: Improved Unlearning-Utility Trade-off with Public Data
ALU uses public data to suppress unlearning cost quadratically while characterizing distribution mismatch effects, enabling mass unlearning with maintained utility.
-
Information as Maximum-Caliber Deviation: A bridge between Integrated Information Theory and the Free Energy Principle
Information defined as maximum-caliber deviation derives IIT 3.0 cause-effect repertoires from constrained entropy maximization and equates to prediction error under CLT and LDT.