EGLR adds a deterministic layer-recursion axis gated by entropy that is complementary to temperature sampling, raising joint oracle accuracy on MATH-500 from 83.4% to 91.6% for a 3B model.
ISBN 979-8-89176-380-7
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
Question generation produces a hidden-state signal that predicts final correctness before the answer is produced, yet gating interventions based on that signal do not reliably improve trajectories.
citing papers explorer
-
Entropy-Gated Latent Recursion
EGLR adds a deterministic layer-recursion axis gated by entropy that is complementary to temperature sampling, raising joint oracle accuracy on MATH-500 from 83.4% to 91.6% for a 3B model.
-
What Am I Missing? Question-Answering as Hidden State Probing
Question generation produces a hidden-state signal that predicts final correctness before the answer is produced, yet gating interventions based on that signal do not reliably improve trajectories.