SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2602.02493 · 2026

PixelGen: Improving Pixel Diffusion with Perceptual Supervision

Shiliang Zhang, Zehong Ma, Ruihan Xu

arXiv · PDF · Open in the Atlas

Code that ran

8 of 11 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 8 have no contract verdict; 0 failures of our instrument; 3 no recorded run), measured 2026-11-21; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2602.02493.json · badge: /api/ran/2602.02493.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
Zehong-Ma/PixelGen community
Papers with Code archive
8 ran · 3 no recorded run
FunctionStatusWhere it lives
Attention Ran
ran on a synthesized input
Zehong-Ma/PixelGen/src/models/transformer/JiT.py
pointer only (licence: NONE) · get_code("cfd8d30a6aecb8ba")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Zehong-Ma/PixelGen when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

BottleneckPatchEmbed Ran
ran on a synthesized input
Zehong-Ma/PixelGen/src/models/transformer/JiT.py
pointer only (licence: NONE) · get_code("da2a72e8e5d467ac")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Zehong-Ma/PixelGen when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

FinalLayer Ran
ran on a synthesized input
Zehong-Ma/PixelGen/src/models/transformer/JiT.py
pointer only (licence: NONE) · get_code("fc4b1cddc56752ce")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Zehong-Ma/PixelGen when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

LabelEmbedder Ran
ran on a synthesized input
Zehong-Ma/PixelGen/src/models/transformer/JiT.py
pointer only (licence: NONE) · get_code("2f4663cf9f632886")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Zehong-Ma/PixelGen when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

PatchEmbed Ran
ran on a synthesized input
Zehong-Ma/PixelGen/src/models/transformer/JiT.py
pointer only (licence: NONE) · get_code("5539727cbcfaae0a")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Zehong-Ma/PixelGen when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

RMSNorm Ran
ran on a synthesized input
Zehong-Ma/PixelGen/src/models/transformer/JiT.py
pointer only (licence: NONE) · get_code("84963760feac9e7d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Zehong-Ma/PixelGen when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

SwiGLUFFN Ran
ran on a synthesized input
Zehong-Ma/PixelGen/src/models/transformer/JiT.py
pointer only (licence: NONE) · get_code("e99f296d2c7d2cfd")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Zehong-Ma/PixelGen when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

TimestepEmbedder Ran
ran on a synthesized input
Zehong-Ma/PixelGen/src/models/transformer/JiT.py
pointer only (licence: NONE) · get_code("b6702ea97cf0e309")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Zehong-Ma/PixelGen when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

JiT Not yet run
unverified (harvested, no recorded run)
Zehong-Ma/PixelGen/src/models/transformer/JiT.py
pointer only (licence: NONE) · get_code("c02e573246876255")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Zehong-Ma/PixelGen when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

JiTBlock Not yet run
unverified (harvested, no recorded run)
Zehong-Ma/PixelGen/src/models/transformer/JiT.py
pointer only (licence: NONE) · get_code("930f46087cc5be89")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Zehong-Ma/PixelGen when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

VisionRotaryEmbeddingFast Not yet run
unverified (harvested, no recorded run)
Zehong-Ma/PixelGen/src/models/transformer/JiT.py
pointer only (licence: NONE) · get_code("658e3e322c0668b6")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Zehong-Ma/PixelGen when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

Also present in other papers' code

Of 11 samples lifted from this paper's linked repositories, 11 carry a function identity and 0 do not. Of the 11 distinct functions with an identity, 7 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 7 later) and 4 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
TimestepEmbedder
brick of this paper's assembly: model
24 other papers
20 dated earlier, 4 later · 20 only as an assembly brick
2303.14389 (2023-03)
Masked Diffusion Transformer is a Strong Image Synthesizer
2305.13311 (2023-05)
VDT: General-purpose Video Diffusion Transformers via Mask Modeling
2401.08741 (2024-01)
Fixed Point Diffusion Models
and 21 more not named here
SwiGLUFFN
brick of this paper's assembly: model
2 other papers
2 later · 1 only as an assembly brick
2603.14366 (2026-03)
Representation Alignment for Just Image Transformers is not Easier than You Think
2608.01298 (2026-08)
UDT: Reconciling U-Nets and Diffusion Transformers with Data-Adaptive Token Reduction
Attention
brick of this paper's assembly: model
1 other paper
1 later
2603.14366 (2026-03)
Representation Alignment for Just Image Transformers is not Easier than You Think
BottleneckPatchEmbed
brick of this paper's assembly: model
1 other paper
1 later
2603.14366 (2026-03)
Representation Alignment for Just Image Transformers is not Easier than You Think
JiTBlock
brick of this paper's assembly: model
1 other paper
1 later
2603.14366 (2026-03)
Representation Alignment for Just Image Transformers is not Easier than You Think
LabelEmbedder
brick of this paper's assembly: model
1 other paper
1 later
2603.14366 (2026-03)
Representation Alignment for Just Image Transformers is not Easier than You Think
RMSNorm
brick of this paper's assembly: model
1 other paper
1 later
2603.14366 (2026-03)
Representation Alignment for Just Image Transformers is not Easier than You Think

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-21.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Pixel diffusion generates images directly in pixel space, avoiding the VAE artifacts and representational bottlenecks of two-stage latent diffusion. Recent JiT further simplifies pixel diffusion with x-prediction, where the model predicts clean images rather than velocity. However, the standard pixel-wise diffusion loss treats all pixels equally, spending model capacity to perceptually insignificant signals and often leading to blurry samples. We propose PixelGen, an end-to-end pixel diffusion framework that augments x-prediction with perceptual supervision. Specifically, PixelGen introduces two complementary perceptual losses on top of x-prediction: an LPIPS loss for local textures and a P-DINO loss for global semantics. To preserve sample coverage, PixelGen further proposes a noise-gating strategy that applies these losses only at lower-noise timesteps. On ImageNet-256 without classifier-free guidance, PixelGen achieves an FID of 5.11 in 80 training epochs, surpassing the latent diffusion baselines. Moreover, Pix-elGen scales efficiently to text-to-image generation, reaching a GenEval score of 0.79 with only 6 days of training on 8×H800 GPUs. These results show that perceptual supervision substantially narrows the gap between pixel and latent diffusion while preserving a simple one-stage pipeline. Codes are available at https://github.com/Zehong-Ma/PixelGen. Diffusion models [1-3] have achieved remarkable success in high-fidelity image generation, offering exceptional quality and diversity. Research in this field generally follows two main directions: latent diffusion and pixel diffusion. Latent diffusion models [4-7] split generation into two stages. As illustrated in Fig. 2(a), a VAE first compresses images into a latent space, and a diffusion model then performs denoising in that space. The performance of latent diffusion is largely constrained by the VAE, where the reconstruction quality limits the upper bound of generation, and the learned latent distribution further affects the convergence of diffusion training [8,9]. These works have shown that VAEs introduce low-level artifacts and representational bottlenecks for latent diffusion models. Pixel diffusion models avoid these limitations by modeling raw pixels directly. This end-to-end pipeline removes the need for latent representations and eliminates VAE-induced artifacts. However, Preprint.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2602.02493")
get_code_for_paper("2602.02493")
have("2602.02493")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2602.02493.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2602.02493.svg)](https://syntology.ai/paper/2602.02493)

Connect an agent — have() is free.