Looped LLMs converge to distinct cyclic fixed points per layer, repeating feedforward-style inference stages across recurrences.
Pay attention to attention distribution: A new lo- cal lipschitz bound for transformers
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.LG 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
A data-parameter correspondence unifies data-centric and parameter-centric LLM optimizations as dual geometric operations on the statistical manifold via Fisher-Rao metric and Legendre duality.
citing papers explorer
-
A Mechanistic Analysis of Looped Reasoning Language Models
Looped LLMs converge to distinct cyclic fixed points per layer, repeating feedforward-style inference stages across recurrences.
-
Towards a Data-Parameter Correspondence for LLMs: A Preliminary Discussion
A data-parameter correspondence unifies data-centric and parameter-centric LLM optimizations as dual geometric operations on the statistical manifold via Fisher-Rao metric and Legendre duality.