Papers › Remasking Discrete Diffusion Models with Inference-Time Scaling

Remasking Discrete Diffusion Models with Inference-Time Scaling

1 Mar 2025arXiv:2503.00307archive 2025-07-28

Guanghan Wang, Yair Schiff, Subham Sekhar Sahoo, Volodymyr Kuleshov

Part of the success of diffusion models stems from their ability to perform iterative refinement, i.e., repeatedly correcting outputs during generation. However, modern masked discrete diffusion lacks this capability: when a token is generated, it cannot be updated again, even when it introduces an error. Here, we address this limitation by introducing the remasking diffusion model (ReMDM) sampler, a method that can be applied to pretrained masked diffusion models in a principled way and that is derived from a discrete diffusion model with a custom remasking backward process. Most interestingly, ReMDM endows discrete diffusion with a form of inference-time compute scaling. By increasing the number of sampling steps, ReMDM generates natural language outputs that approach the quality of autoregressive models, whereas when the computation budget is limited, ReMDM better maintains quality. ReMDM also improves sample quality of masked diffusion models for discretized images, and in scientific domains such as molecule design, ReMDM facilitates diffusion guidance and pushes the Pareto frontier of controllability relative to classical masking and uniform noise diffusion. We provide the code along with a blog post on the project page: https://remdm.github.io.

PaperPDFCodeCode Syntology ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2503.00307")

Code

Syntology Ran 13 of 17 code samples harvested from 1 repository linked to this paper; 4 have no recorded run. Of those that ran: 2 ran · our draft was wrong; 1 ran · fixture could not drive it; 10 ran with no contract checked.

By repository: community (archive-listed): 17 samples from 1 repository, 13 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

kuleshov-group/remdm mentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

17 samples harvested; 13 ran; 0 honoured the contract we drafted; 4 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

2ran · our draft was wrong
1ran · fixture could not drive it
10ran
4unverified

Licence: 0 of the 17 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from kuleshov-group/remdm. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

BPD kuleshov-group/remdm/diffusion.py community (archive-listed) ran Apache-2.0 (permissive) · 8a96d6ee76d899fb · report
CosineNoise kuleshov-group/remdm/diffusion.py community (archive-listed) ran fingerprinted Apache-2.0 (permissive) · 2252e72b7e6ff199 · report
CosineSqrNoise kuleshov-group/remdm/diffusion.py community (archive-listed) ran fingerprinted Apache-2.0 (permissive) · f3c78ccd48148489 · report
FaultTolerantDistributedSampler kuleshov-group/remdm/diffusion.py community (archive-listed) ran Apache-2.0 (permissive) · a8eb0d121c7ecbcc · report
GeometricNoise kuleshov-group/remdm/diffusion.py community (archive-listed) ran fingerprinted Apache-2.0 (permissive) · 06f92d50089269a3 · report
Linear kuleshov-group/remdm/diffusion.py community (archive-listed) ran fingerprinted Apache-2.0 (permissive) · 23091ee635899def · report
LogLinearNoise kuleshov-group/remdm/diffusion.py community (archive-listed) ran fingerprinted Apache-2.0 (permissive) · 721fda60a2762576 · report
Loss kuleshov-group/remdm/diffusion.py community (archive-listed) ran Apache-2.0 (permissive) · f0c0a2810b675abb · report
Perplexity kuleshov-group/remdm/diffusion.py community (archive-listed) ran Apache-2.0 (permissive) · e3f789992ecfe962 · report
RandomFaultTolerantSampler kuleshov-group/remdm/diffusion.py community (archive-listed) ran Apache-2.0 (permissive) · ca44ec1d6c02bb7d · report
_sample_categorical kuleshov-group/remdm/diffusion.py community (archive-listed) ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 1c47525224ba228c · report
_unsqueeze kuleshov-group/remdm/diffusion.py community (archive-listed) ran · fixture could not drive it fingerprinted Apache-2.0 (permissive) · 30c96381b7f09da5 · report
get_noise kuleshov-group/remdm/diffusion.py community (archive-listed) ran · our draft was wrong Apache-2.0 (permissive) · 2174721db8d9cdb6 · report
Diffusion kuleshov-group/remdm/diffusion.py community (archive-listed) unverified Apache-2.0 (permissive) · 2c650372d2d0b176 · report
NLL kuleshov-group/remdm/diffusion.py community (archive-listed) unverified Apache-2.0 (permissive) · c7461f97befe2293 · report
Noise kuleshov-group/remdm/diffusion.py community (archive-listed) unverified Apache-2.0 (permissive) · 4360fa80bd3c0ebd · report
print_nans kuleshov-group/remdm/diffusion.py community (archive-listed) unverified Apache-2.0 (permissive) · 1665eba64e29d70c · report

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Diffusion

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections