Papers › Why Masking Diffusion Works: Condition on the Jump Schedule for Improved Discrete Diffusion

Why Masking Diffusion Works: Condition on the Jump Schedule for Improved Discrete Diffusion

10 Jun 2025arXiv:2506.08316archive 2025-07-28

Alan N. Amin, Nate Gruver, Andrew Gordon Wilson

Discrete diffusion models, like continuous diffusion models, generate high-quality samples by gradually undoing noise applied to datapoints with a Markov process. Gradual generation in theory comes with many conceptual benefits; for example, inductive biases can be incorporated into the noising Markov process, and access to improved sampling algorithms. In practice, however, the consistently best performing discrete diffusion model is, surprisingly, masking diffusion, which does not denoise gradually. Here we explain the superior performance of masking diffusion by noting that it makes use of a fundamental difference between continuous and discrete Markov processes: discrete Markov processes evolve by discontinuous jumps at a fixed rate and, unlike other discrete diffusion models, masking diffusion builds in the known distribution of jump times and only learns where to jump to. We show that we can similarly bake in the known distribution of jump times into any discrete diffusion model. The resulting models - schedule-conditioned discrete diffusion (SCUD) - generalize classical discrete diffusion and masking diffusion. By applying SCUD to models with noising processes that incorporate inductive biases on images, text, and protein data, we build models that outperform masking.

PaperPDFCodeCode Syntology ranMethod as code

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

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

Syntology holds a reference implementation of SCUD (schedule-conditioned diffusion), generated from the paper text (not the authors' code), at level 2 on Syntology's ladder: V2 “two implementations GENERATED from the paper's signature and spec agree numerically (1e-6) on fixture inputs” “neither of the two ever read the authors' own code, so V2 measures how unambiguously the paper specifies the method, not fidelity to the authors' implementation (measured 2026-09-11, commit 75eaa77: cross_checked_by is a second LLM on 2,628 of 2,831 served samples)” (quoted from Syntology's server card). No rung compares this code against the authors' implementation. For agents: get_reference_implementation(name='SCUD (schedule-conditioned diffusion)') (how to connect); the free check have('2506.08316') needs no token.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 16 code samples harvested from 1 repository linked to this paper: 10 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 8 where Syntology's instrument failed) · 6 unverified. Where Syntology's instrument failed: 7 where our contract draft was wrong; 1 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 16 samples from 1 repository, 10 ran (2 with no instrument failure; 8 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

No sample is linked to this paper from 1 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

alannawzadamin/scud officialmentioned in papermentioned on GitHubpytorchMITSyntology: 16 samples, 10 ran (8 instrument-failed) · 6 unverified 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

16 samples harvested: 10 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 8 where Syntology's instrument failed) · 6 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · honoured contract
7ran · our draft was wrong
1ran; the contract check could not run
1ran
6unverified

Licence: 0 of the 16 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 alannawzadamin/scud. “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. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (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 are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. 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 six 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; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. 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.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

alannawzadamin/scud official repository 16 samples, 10 ran (8 instrument-failed) · 6 unverified
convert_to_probs scud/scud.py ran · our draft was wrong fingerprinted MIT (permissive) · 0325c917fceb7495 · report fixture: signature x_0:Tensor/2/int, num_classes:int/0/int · returned Tensor · level recorded 2026-09-22
get_a_b_func_sc scud/scud.py ran fingerprinted MIT (permissive) · 340e4d9099615dcc · report fixture: signature K:Tensor/2/float, p0:Tensor/1/float, precompute_mis:NoneType/0/?, second_eval:NoneType/0/? · returned tuple · level recorded 2026-09-22
get_gif scud/scud.py ran · our draft was wrong MIT (permissive) · dc3cbb6c4c03d2c2 · report fixture: signature sample_x:Tensor/4/float, sample_a:Tensor/2/float, model:callable/0/?, gen_trans_step:int/0/int, batch_size:int/0/int · returned tuple · level recorded 2026-09-22
get_inf_gen scud/scud.py ran; the contract check could not run (DEP_MISSING) MIT (permissive) · 708ec5714fd16adf · report fixture: signature forward_kwargs:dict/0/?, num_classes:int/0/int · returned Tensor · level recorded 2026-09-22
get_text in 18 papers scud/scud.py ran · our draft was wrong MIT (permissive) · 171b7d239afea3df · report fixture: signature sample_x:Tensor/3/float, sample_a:Tensor/2/float, model:callable/0/?, gen_trans_step:int/0/int, batch_size:int/0/int, tokenizer:MinimalTokenizer/0/? · returned tuple · level recorded 2026-09-24
hash_matrix scud/scud.py ran · our draft was wrong MIT (permissive) · 2ba39f54891a0f2e · report fixture: signature matrix:Tensor/2/float · returned str · level recorded 2026-09-22
kls scud/scud.py ran · our draft was wrong fingerprinted MIT (permissive) · e8d679e578149494 · report fixture: signature dist1:Tensor/2/float, dist2:Tensor/2/float, eps:NoneType/0/? · returned Tensor · level recorded 2026-09-22
newton_root_finder scud/scud.py ran · our draft was wrong MIT (permissive) · 828ffbe52bb6c5b1 · report fixture: signature func:callable/0/?, x0:Tensor/0/float, ts:Tensor/1/float, min_x:Tensor/0/float, max_iter:int/0/int, tol:float/0/float, print_:bool/0/bool · returned Tensor · level recorded 2026-09-22
sample_n_transitions_cont scud/scud.py ran · our draft was wrong MIT (permissive) · 8bd5456946e5396e · report fixture: signature log_alpha:callable/0/?, batch_size:int/0/int, times:Tensor/1/float · returned Tensor · level recorded 2026-09-22
try_load scud/scud.py ran · honoured contract MIT (permissive) · 113fa009b87737b7 · report fixture: signature func:callable/0/?, fname:str/0/str, path:str/0/str · returned Tensor · level recorded 2026-09-22
ContinuousTimeDiffusion scud/scud.py unverified MIT (permissive) · 350372f89ed37c5f · report
DiffusionTrainer scud/scud.py unverified MIT (permissive) · a47877ec65e2b4e1 · report
SCUD scud/scud.py unverified MIT (permissive) · 90974b46a044c53a · report
get_a_b_func_cont scud/scud.py unverified MIT (permissive) · 7245f56fde77ea57 · report
get_a_b_func_mi scud/scud.py unverified MIT (permissive) · f0b1e6873435a4df · report
get_betas in 9 papers scud/scud.py unverified MIT (permissive) · 00ed359ed5a206e4 · report

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. For a generated assembly, an independent peer is a second implementation generated separately from the same paper text; neither implementation read the authors' code. At level 2 the two agreed numerically on the fixture's inputs; level 3 is recorded when property tests derived from the paper also passed on both, and each level 3 card prints how many its report lists. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it. A generated assembly's label was set when its code was generated for the method, and no model judged it; its card prints it as "by construction" or "not judged by a model". Where an assembly's phrase has no constructed figure, a brick that ran is not among the harvested samples this build reads, and whether a brick's run only constructed an object is read from that brick's harvested sample, so the figure was not counted; it does not say there were none.

2 assemblies on this paper: 1 harvested, 1 generated.

generated Assembly for SCUD (schedule-conditioned diffusion) (graph method: SCUD (schedule-conditioned diffusion)), generated from the paper text, not the repository's code. 1 brick; individually on a fixture: 1 ran (1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified. Ran end-to-end against an independent peer (level 2).

label matches by construction: the code was generated for this method (not judged by a model)

  1. multi_step_denoise_categorical entry · level 2 · ran · 1e8f8d62481195d2077975d021168d3dcafc11f1cabdfec361aa07e313534612

harvested Assembly for SCUD (schedule-conditioned diffusion) (graph method: SCUD (schedule-conditioned diffusion)). 16 bricks; individually on a fixture: 10 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 8 where Syntology's instrument failed) · 6 unverified. The assembly has not been run end-to-end.

  1. SCUD entry · level 0 · unverified · 90974b46a044c53a
  2. ContinuousTimeDiffusion helper · level 0 · unverified · 350372f89ed37c5f
  3. convert_to_probs helper · level 1 · ran · our draft was wrong · 0325c917fceb7495
  4. get_inf_gen helper · level 1 · ran; the contract check could not run (DEP_MISSING) · 708ec5714fd16adf
  5. kls helper · level 1 · ran · our draft was wrong · e8d679e578149494
  6. sample_n_transitions_cont helper · level 1 · ran · our draft was wrong · 8bd5456946e5396e
  7. DiffusionTrainer helper · level 0 · unverified · a47877ec65e2b4e1
  8. get_betas scheduler · level 0 · unverified · 00ed359ed5a206e4
  9. get_gif helper · level 1 · ran · our draft was wrong · dc3cbb6c4c03d2c2
  10. get_text helper · level 1 · ran · our draft was wrong · 171b7d239afea3df
  11. get_a_b_func_mi helper · level 0 · unverified · f0b1e6873435a4df
  12. get_a_b_func_cont helper · level 0 · unverified · 7245f56fde77ea57
  13. get_a_b_func_sc helper · level 1 · ran · 340e4d9099615dcc
  14. hash_matrix helper · level 1 · ran · our draft was wrong · 2ba39f54891a0f2e
  15. newton_root_finder helper · level 1 · ran · our draft was wrong · 828ffbe52bb6c5b1
  16. try_load helper · level 1 · ran · honoured contract · 113fa009b87737b7

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2506.08316") returns these assemblies with each brick's role and wiring (how to connect).

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