Showing cs.CLShow all
2 papers · 1 filter
cs.CL2026
Re-Mask and Redirect: Exploiting Denoising Irreversibility in Diffusion Language Models
Arth Singh
Safety alignment in diffusion language models (dLLMs) relies on a single load-bearing assumption: that committed tokens are permanent. We show that violating this assumption, by re…
cs.CL2026
EMA Is Not All You Need: Mapping the Boundary Between Structure and Content in Recurrent Context
Arth Singh
What exactly do efficient sequence models gain over simple temporal averaging? We use exponential moving average (EMA) traces, the simplest recurrent context (no gating, no content…