pith. sign in

Universal Transformers

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

fields

cs.LG 1

years

2026 1

verdicts

CONDITIONAL 1

representative citing papers

Relational Preference Encoding in Looped Transformer Internal States

cs.LG · 2026-04-10 · conditional · novelty 6.0

Looped transformer hidden states encode preferences relationally via pairwise differences rather than independent pointwise classification, with the evaluator acting as an internal consistency probe on the model's own value system.

citing papers explorer

Showing 1 of 1 citing paper.

  • Relational Preference Encoding in Looped Transformer Internal States cs.LG · 2026-04-10 · conditional · none · ref 5

    Looped transformer hidden states encode preferences relationally via pairwise differences rather than independent pointwise classification, with the evaluator acting as an internal consistency probe on the model's own value system.