Tag
A research paper analyzing what iterated self-feeding probes of language models actually measure, distinguishing model-dependent signals from construction artifacts using a ring of resampled token cells and common random numbers coupling.
This paper presents a polynomial-time algorithm for learning the structure of a Gaussian graphical model from a single trajectory of Glauber dynamics, with a trajectory-length guarantee that does not depend on the mixing time.
This paper analyzes the global distributional behavior induced by iterative masked-token resampling in masked language models using Glauber dynamics. It introduces a rectangle test for incompatibility, establishes mixing time bounds, and empirically demonstrates phase transitions and metastable semantic basins.