← Search

Scott W Viteri

1 accepted papers

2026

Markovian Transformers for Informative Language Modeling

ICLR 2026poster

Chain-of-Thought (CoT) reasoning often fails to faithfully reflect a language model's underlying decision process. We address this by introducing a \emph{Markovian} language model framework with an autoencoder-style \emph{reasoning bottleneck}: it creates a text-based bottleneck where CoT serves as…

Cited by 0SourcecodeScholar