Papers › Towards Stabilized and Efficient Diffusion Transformers through Long-Skip-Connections...

Towards Stabilized and Efficient Diffusion Transformers through Long-Skip-Connections with Spectral Constraints

26 Nov 2024arXiv:2411.17616archive 2025-07-28

Guanjie Chen, Xinyu Zhao, Yucheng Zhou, Xiaoye Qu, Tianlong Chen, Yu Cheng

Diffusion Transformers (DiT) have emerged as a powerful architecture for image and video generation, offering superior quality and scalability. However, their practical application suffers from inherent dynamic feature instability, leading to error amplification during cached inference. Through systematic analysis, we identify the absence of long-range feature preservation mechanisms as the root cause of unstable feature propagation and perturbation sensitivity. To this end, we propose Skip-DiT, a novel DiT variant enhanced with Long-Skip-Connections (LSCs) - the key efficiency component in U-Nets. Theoretical spectral norm and visualization analysis demonstrate how LSCs stabilize feature dynamics. Skip-DiT architecture and its stabilized dynamic feature enable an efficient statical caching mechanism that reuses deep features across timesteps while updating shallow components. Extensive experiments across image and video generation tasks demonstrate that Skip-DiT achieves: (1) 4.4 times training acceleration and faster convergence, (2) 1.5-2 times inference acceleration without quality loss and high fidelity to original output, outperforming existing DiT caching methods across various quantitative metrics. Our findings establish long-skip connections as critical architectural components for training stable and efficient diffusion transformers.

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="2411.17616")

Code

Syntology Ran 4 of 10 code samples harvested from 1 repository linked to this paper; 6 have no recorded run. Of those that ran: 1 ran · honoured contract; 3 ran · our draft was wrong.

By repository: official repository: 10 samples from 1 repository, 4 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

opensparsellms/skip-dit officialmentioned in papermentioned on GitHubpytorchApache-2.0 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

10 samples harvested; 4 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
3ran · our draft was wrong
6unverified

Licence: 0 of the 10 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 opensparsellms/skip-dit. “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.

create_npz_from_sample_folder opensparsellms/skip-dit/class-to-image/sample_ddp.py official repository ran · our draft was wrong Apache-2.0 (permissive) · 7b21a01ae77703a3 · report
get_2d_sincos_pos_embed opensparsellms/skip-dit/class-to-image/models.py official repository ran · honoured contract Apache-2.0 (permissive) · c92c27c924b517e8 · report
modulate opensparsellms/skip-dit/class-to-image/models.py official repository ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 03310bba324ae4fb · report
zero_module opensparsellms/skip-dit/class-to-video/models/utils.py official repository ran · our draft was wrong Apache-2.0 (permissive) · 129b804760b3115f · report
checkpoint opensparsellms/skip-dit/class-to-video/models/utils.py official repository unverified Apache-2.0 (permissive) · 6fe195fcfd10e259 · report
download_model opensparsellms/skip-dit/class-to-image/download.py official repository unverified Apache-2.0 (permissive) · 6a0d5ecd441905cc · report
find_model opensparsellms/skip-dit/class-to-image/download.py official repository unverified Apache-2.0 (permissive) · 29947a0a94157558 · report
get_1d_sincos_temp_embed opensparsellms/skip-dit/class-to-video/models/latte.py official repository unverified Apache-2.0 (permissive) · c8dc321d086803c8 · report
get_2d_sincos_pos_embed_from_grid opensparsellms/skip-dit/class-to-image/models.py official repository unverified Apache-2.0 (permissive) · 665d8a4e8f673a4c · report
timestep_embedding opensparsellms/skip-dit/class-to-video/models/utils.py official repository unverified Apache-2.0 (permissive) · 319f00691ca1b97c · report

Tasks

DenoisingImage GenerationVideo Generation

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