SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2408.05710 · ECCV · 2024

Efficient Diffusion Transformer with Step-wise Dynamic Attention Mediators

Gao Huang, Shiji Song, Xiu Li, Yuhui Yuan, Yizeng Han, Yifan Pu, Dongchen Han, Zhuofan Xia, Jiayi Guo, Qixiu Li, Ji Li, Duo Li

arXiv · PDF · Open in the Atlas

Code that ran

12 of 15 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (1 of them constructed an object rather than computing a result; 1 honoured the contract we drafted, 0 violated it and 1 has no contract verdict; 10 failures of our instrument; 3 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; 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/2408.05710.json · badge: /api/ran/2408.05710.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
leaplabthu/attention-mediators canonical (classifier label)
classifier basis: census
11 ran · 3 no recorded run
LeapLabTHU/Attention-Mediators community
Papers with Code archive
1 ran · 0 no recorded run
FunctionStatusWhere it lives
AttentionWithMediators Ran
ran on a synthesized input · constructed an object; no result was computed
LeapLabTHU/Attention-Mediators/attention_mediator/models.py
pointer only (licence: NONE) · get_code("66c4a4e6a1c9f9e5")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, LeapLabTHU/Attention-Mediators 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.

array2grid Ran
ran; our contract draft was wrong, not the code
leaplabthu/attention-mediators/attention_mediator/wandb_utils.py
pointer only (licence: NONE) · get_code("4a9faf5d974dbff7")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, leaplabthu/attention-mediators 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.

center_crop_arr Ran
ran; our contract draft was wrong, not the code
leaplabthu/attention-mediators/attention_mediator/train_256to512.py
pointer only (licence: NONE) · get_code("95bcfe1a68b3578c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, leaplabthu/attention-mediators 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.

create_npz_from_sample_folder Ran
ran; our contract draft was wrong, not the code
leaplabthu/attention-mediators/attention_mediator/sample_ddp.py
pointer only (licence: NONE) · get_code("7b21a01ae77703a3")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, leaplabthu/attention-mediators 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.

estimated_time Ran
ran; our contract draft was wrong, not the code
leaplabthu/attention-mediators/attention_mediator/train_256to512.py
pointer only (licence: NONE) · get_code("2d3b1f803fa682d9")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, leaplabthu/attention-mediators 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.

expand_t_like_x Ran
ran; our contract draft was wrong, not the code
leaplabthu/attention-mediators/attention_mediator/transport/path.py
pointer only (licence: NONE); the stored copy get_code serves, from poppuppy/secant-expectation, is pointer only (licence: NONE) · get_code("6de57ffd4ee7fee7")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, leaplabthu/attention-mediators 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.

generate_run_id Ran
ran; our contract draft was wrong, not the code
leaplabthu/attention-mediators/attention_mediator/wandb_utils.py
pointer only (licence: NONE) · get_code("661fd94e23de34b1")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, leaplabthu/attention-mediators 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.

get_2d_sincos_pos_embed Ran
ran, honoured the contract we drafted
leaplabthu/attention-mediators/attention_mediator/models.py
pointer only (licence: NONE) · get_code("c92c27c924b517e8")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, leaplabthu/attention-mediators 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.

log_state Ran
ran; our contract draft was wrong, not the code
leaplabthu/attention-mediators/attention_mediator/transport/utils.py
pointer only (licence: NONE) · get_code("e0c9eec71fe93415")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, leaplabthu/attention-mediators 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.

mean_flat Ran
ran; our contract draft was wrong, not the code
leaplabthu/attention-mediators/attention_mediator/transport/utils.py
pointer only (licence: NONE); the stored copy get_code serves, from poppuppy/secant-expectation, is pointer only (licence: NONE) · get_code("b883daa24762cb05")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, leaplabthu/attention-mediators 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.

modulate Ran
ran; our contract draft was wrong, not the code
leaplabthu/attention-mediators/attention_mediator/models.py
pointer only (licence: NONE) · get_code("03310bba324ae4fb")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, leaplabthu/attention-mediators 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.

namespace_to_dict Ran
ran; our contract draft was wrong, not the code
leaplabthu/attention-mediators/attention_mediator/wandb_utils.py
pointer only (licence: NONE) · get_code("06a3ad042b36b928")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, leaplabthu/attention-mediators 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.

create_logger Not yet run
unverified (harvested, no recorded run)
leaplabthu/attention-mediators/attention_mediator/train_256to512.py
pointer only (licence: NONE); the stored copy get_code serves, from quandao10/slct, is pointer only (licence: NONE) · get_code("8382f7e48cc8a309")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, leaplabthu/attention-mediators 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.

get_2d_sincos_pos_embed_from_grid Not yet run
unverified (harvested, no recorded run)
leaplabthu/attention-mediators/attention_mediator/models.py
pointer only (licence: NONE) · get_code("665d8a4e8f673a4c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, leaplabthu/attention-mediators 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.

none_or_str Not yet run
unverified (harvested, no recorded run)
leaplabthu/attention-mediators/attention_mediator/train_utils.py
pointer only (licence: NONE) · get_code("2fc6fdb85f13dc41")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, leaplabthu/attention-mediators 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 15 samples lifted from this paper's linked repositories, 15 carry a function identity and 0 do not. Of the 15 distinct functions with an identity, 12 are also present in other papers whose code we hold (12 with at least one paper dated earlier, 5 the same month, 9 later) and 3 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
get_2d_sincos_pos_embed_from_grid 190 other papers
104 dated earlier, 6 the same month, 79 later, 1 same year, order not determinable · 65 only as an assembly brick
1807.10221 (2018-07)
Unified Perceptual Parsing for Scene Understanding
2105.12723 (2021-05)
Nested Hierarchical Transformer: Towards Accurate, Data-Efficient and Interpretable Visual Understanding
2111.06377 (2021-11)
Masked Autoencoders Are Scalable Vision Learners
and 187 more not named here
modulate 101 other papers
40 dated earlier, 2 the same month, 58 later, 1 undated · 29 only as an assembly brick
2202.04200 (2022-02)
MaskGIT: Masked Generative Image Transformer
2203.10897 (2022-03)
Unified Multivariate Gaussian Mixture for Efficient Neural Image Compression
2212.09748 (2022-12)
Scalable Diffusion Models with Transformers
and 98 more not named here
get_2d_sincos_pos_embed 69 other papers
29 dated earlier, 1 the same month, 39 later · 12 only as an assembly brick
2203.10897 (2022-03)
Unified Multivariate Gaussian Mixture for Efficient Neural Image Compression
2212.09748 (2022-12)
Scalable Diffusion Models with Transformers
2303.14389 (2023-03)
Masked Diffusion Transformer is a Strong Image Synthesizer
and 66 more not named here
create_npz_from_sample_folder 23 other papers
7 dated earlier, 16 later
2401.08740 (2024-01)
SiT: Exploring Flow and Diffusion-based Generative Models with Scalable Interpolant Transformers
2402.14167 (2024-02)
T-Stitch: Accelerating Sampling in Pre-Trained Diffusion Models with Trajectory Stitching
2403.09176 (2024-03)
Switch Diffusion Transformer: Synergizing Denoising Tasks with Sparse Mixture-of-Experts
and 20 more not named here
none_or_str 16 other papers
7 dated earlier, 1 the same month, 8 later
1810.01845 (2018-10)
Task-Oriented Hand Motion Retargeting for Dexterous Manipulation Imitation
1909.10893 (2019-09)
Recurrent Independent Mechanisms
2109.11251 (2021-09)
Trust Region Policy Optimisation in Multi-Agent Reinforcement Learning
and 13 more not named here
create_logger 14 other papers
4 dated earlier, 1 the same month, 9 later
2212.09748 (2022-12)
Scalable Diffusion Models with Transformers
2405.20337 (2024-05)
OccSora: 4D Occupancy Generation Models as World Simulators for Autonomous Driving
2406.01733 (2024-06)
Learning-to-Cache: Accelerating Diffusion Transformer via Layer Caching
and 11 more not named here
expand_t_like_x 5 other papers
2 dated earlier, 3 later
2401.08740 (2024-01)
SiT: Exploring Flow and Diffusion-based Generative Models with Scalable Interpolant Transformers
2403.13802 (2024-03)
ZigMa: A DiT-style Zigzag Mamba Diffusion Model
2409.18128 (2024-09)
FlowTurbo: Towards Real-time Flow-Based Image Generation with Velocity Refiner
and 2 more not named here
mean_flat 4 other papers
1 dated earlier, 3 later
2401.08740 (2024-01)
SiT: Exploring Flow and Diffusion-based Generative Models with Scalable Interpolant Transformers
2409.18128 (2024-09)
FlowTurbo: Towards Real-time Flow-Based Image Generation with Velocity Refiner
2503.21758 (2025-03)
Lumina-Image 2.0: A Unified and Efficient Image Generative Framework
and 1 more not named here
log_state 3 other papers
1 dated earlier, 2 later
2401.08740 (2024-01)
SiT: Exploring Flow and Diffusion-based Generative Models with Scalable Interpolant Transformers
2409.18128 (2024-09)
FlowTurbo: Towards Real-time Flow-Based Image Generation with Velocity Refiner
2503.21758 (2025-03)
Lumina-Image 2.0: A Unified and Efficient Image Generative Framework
generate_run_id 2 other papers
2 dated earlier
2401.08740 (2024-01)
SiT: Exploring Flow and Diffusion-based Generative Models with Scalable Interpolant Transformers
2403.13802 (2024-03)
ZigMa: A DiT-style Zigzag Mamba Diffusion Model
namespace_to_dict 2 other papers
2 dated earlier
2401.08740 (2024-01)
SiT: Exploring Flow and Diffusion-based Generative Models with Scalable Interpolant Transformers
2403.13802 (2024-03)
ZigMa: A DiT-style Zigzag Mamba Diffusion Model
array2grid 1 other paper
1 dated earlier
2401.08740 (2024-01)
SiT: Exploring Flow and Diffusion-based Generative Models with Scalable Interpolant Transformers

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 2026-11-22.

Repositories linked to this paper

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

Abstract

This paper identifies significant redundancy in the query-key interactions within self-attention mechanisms of diffusion transformer models, particularly during the early stages of denoising diffusion steps. In response to this observation, we present a novel diffusion transformer framework incorporating an additional set of mediator tokens to engage with queries and keys separately. By modulating the number of mediator tokens during the denoising generation phases, our model initiates the denoising process with a precise, non-ambiguous stage and gradually transitions to a phase enriched with detail. Concurrently, integrating mediator tokens simplifies the attention module's complexity to a linear scale, enhancing the efficiency of global attention processes. Additionally, we propose a time-step dynamic mediator token adjustment mechanism that further decreases the required computational FLOPs for generation, simultaneously facilitating the generation of high-quality images within the constraints of varied inference budgets. Extensive experiments demonstrate that the proposed method can improve the generated image quality while also reducing the inference cost of diffusion transformers. When integrated with the recent work SiT, our method achieves a state-of-the-art FID score of 2.01. The source code is available at https://github.com/LeapLabTHU/Attention-Mediators.

For agents

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

get_harvested_code_for_paper("2408.05710")
get_code_for_paper("2408.05710")
have("2408.05710")

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

curl https://syntology.ai/api/ran/2408.05710.json

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

[![Syntology run record](https://syntology.ai/api/ran/2408.05710.svg)](https://syntology.ai/paper/2408.05710)

Connect an agent — have() is free.