Masked diffusion models are equivalent to polynomially-padded PLTs, solve all CoT-augmented transformer problems, and are more efficient than CoT for regular languages due to parallelism.
pnmodNq `1 represents the position within that layer. T 1 then processes a stringwPΣ ˚ of lengthNpadded withN 2 padding symbols as follows. (1)T 1 uses an additional “copy
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
On the Reasoning Abilities of Masked Diffusion Language Models
Masked diffusion models are equivalent to polynomially-padded PLTs, solve all CoT-augmented transformer problems, and are more efficient than CoT for regular languages due to parallelism.