3 papers
cs.LG2026
Privileged Context as Drift in On-Policy Self-Distillation
Ravenor Davion, Nick Rui
On-policy self-distillation (OPSD) trains a language model to match a copy of itself conditioned on privileged context. Existing work varies what privileged context contains and ho…
cs.LG2026
Where's the Plan? Locating Latent Planning in Language Models with Lightweight Mechanistic Interventions
Nicole Ma, Nick Rui
We study planning site formation in language models -- where internal representations of structurally-constrained future tokens form during the forward pass, and whether they causa…
cs.LG2026
Test-Time Meta-Adaptation with Self-Synthesis
Zeyneb N. Kaya, Nick Rui
As strong general reasoners, large language models (LLMs) encounter diverse domains and tasks, where the ability to adapt and self-improve at test time is valuable. We introduce MA…