[KongchangAI]
Conceptremasking

Remasking

An operation in the iterative denoising process of diffusion language models that re-marks committed tokens as undetermined so they have a chance to be re-predicted, used to adjust the generation trajectory.

Timeline (last 90 days)

Nov 17

Remasking methods require additional forward evaluations to progressively replace tokens, incurring substantial computational overhead.

Unverified50%
Nov 13

On the Sudoku task, when re-masking an isolated error while keeping the surrounding context unchanged, the regenerated result is often the same error again.

Unverified50%
Nov 10

Remasking refers to re-marking certain already-committed tokens as undetermined before the next denoising step, giving them a chance to be re-predicted.

Unverified50%
Sep 23

Continuous diffusion language models map discrete tokens to a continuous embedding space, perform the noising and denoising process in that space, and finally decode back to discrete tokens.

Unverified50%

All Facts (4)

Source Articles