Papers › Any-Order GPT as Masked Diffusion Model: Decoupling Formulation and Architecture

Any-Order GPT as Masked Diffusion Model: Decoupling Formulation and Architecture

24 Jun 2025arXiv:2506.19935archive 2025-07-28

Shuchen Xue, Tianyu Xie, Tianyang Hu, Zijin Feng, Jiacheng Sun, Kenji Kawaguchi, Zhenguo Li, Zhi-Ming Ma

Large language models (LLMs) predominantly use autoregressive (AR) approaches, but masked diffusion models (MDMs) are emerging as viable alternatives. A key challenge in comparing AR and MDM paradigms is their typical architectural difference: AR models are often decoder-only, while MDMs have largely been encoder-only. This practice of changing both the modeling paradigm and architecture simultaneously makes direct comparisons unfair, as it's hard to distinguish whether observed differences stem from the paradigm itself or the architectural shift. This research evaluates MDMs within a decoder-only framework to: (1) equitably compare MDM (as Any-Order AR, or AO-AR) and standard AR paradigms. Our investigation suggests that the standard AO-AR objective, which averages over all token permutations, may benefit from refinement, as many permutations appear less informative compared to the language's inherent left-to-right structure. (2) Investigate architectural influences (decoder-only vs. encoder-only) within MDMs. We demonstrate that while encoder-only MDMs model a simpler conditional probability space, decoder-only MDMs can achieve dramatic generation speedups (∼25×) and comparable perplexity with temperature annealing despite modeling a vastly larger space, highlighting key trade-offs. This work thus decouples core paradigm differences from architectural influences, offering insights for future model design. Code is available at https://github.com/scxue/AO-GPT-MDM.

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="2506.19935")

Code

Syntology Ran 8 of 14 code samples harvested from 1 repository linked to this paper; 6 have no recorded run. Of those that ran: 1 ran · honoured contract; 7 ran with no contract checked.

By repository: official repository: 14 samples from 1 repository, 8 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

scxue/ao-gpt-mdm officialmentioned in paperpytorch 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

14 samples harvested; 8 ran; 1 honoured the contract we drafted; 6 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.

1ran · honoured contract
7ran
6unverified

Licence: 14 of the 14 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 scxue/AO-GPT-MDM. “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.

AOGPT scxue/AO-GPT-MDM/model_AOGPT_AdaLN6_NoRep_cond_128_trunc_qknorm.py official repository ran no licence file found · pointer only · 2c1c6667a1ed50a0 · report
CausalSelfAttention scxue/AO-GPT-MDM/model_AOGPT_AdaLN6_NoRep_cond_128_trunc_qknorm.py official repository ran no licence file found · pointer only · 311dd4150ecf301e · report
FinalLayer scxue/AO-GPT-MDM/model_AOGPT_AdaLN6_NoRep_cond_128_trunc_qknorm.py official repository ran no licence file found · pointer only · f3bdc88f4e362aea · report
RMSNorm scxue/AO-GPT-MDM/model_AOGPT_AdaLN6_NoRep_cond_128_trunc_qknorm.py official repository ran fingerprinted no licence file found · pointer only · 7d84ead0fe5b239f · report
lm1b_detokenizer scxue/ao-gpt-mdm/data.py official repository ran fingerprinted no licence file found · pointer only · 576e1b9b9bcae8ff · report
modulate scxue/ao-gpt-mdm/model_AOGPT_AdaLN6_NoRep_cond_128_trunc_qknorm.py official repository ran · honoured contract fingerprinted no licence file found · pointer only · 62fcb3912a967a50 · report
ptb_detokenizer scxue/ao-gpt-mdm/data.py official repository ran fingerprinted no licence file found · pointer only · 9f7fbdbe7ca280e0 · report
wt_detokenizer scxue/ao-gpt-mdm/data.py official repository ran fingerprinted no licence file found · pointer only · 46c10627206a7795 · report
Block scxue/AO-GPT-MDM/model_AOGPT_AdaLN6_NoRep_cond_128_trunc_qknorm.py official repository unverified no licence file found · pointer only · ad3472bf14169653 · report
add_scale_fused scxue/ao-gpt-mdm/model_AOGPT_generate.py official repository unverified no licence file found · pointer only · d3ed17c7a19c2660 · report
gumbel_softmax scxue/ao-gpt-mdm/catsample.py official repository unverified no licence file found · pointer only · a017bf60b761f427 · report
modulate_fused scxue/ao-gpt-mdm/model_AOGPT_generate.py official repository unverified no licence file found · pointer only · ae815ba2af68a57f · report
sample_categorical scxue/ao-gpt-mdm/catsample.py official repository unverified no licence file found · pointer only · 363f8d07ea5498f8 · report
top_p_sampling_from_probs scxue/ao-gpt-mdm/catsample.py official repository unverified no licence file found · pointer only · 7f1c917dfe447509 · report

Tasks

Decoder

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