Papers › Is Attention Better Than Matrix Decomposition?
Is Attention Better Than Matrix Decomposition?
Zhengyang Geng, Meng-Hao Guo, Hongxu Chen, Xia Li, Ke Wei, Zhouchen Lin
As an essential ingredient of modern deep learning, attention mechanism, especially self-attention, plays a vital role in the global correlation discovery. However, is hand-crafted attention irreplaceable when modeling the global context? Our intriguing finding is that self-attention is not better than the matrix decomposition (MD) model developed 20 years ago regarding the performance and computational cost for encoding the long-distance dependencies. We model the global context issue as a low-rank recovery problem and show that its optimization algorithms can help design global information blocks. This paper then proposes a series of Hamburgers, in which we employ the optimization algorithms for solving MDs to factorize the input representations into sub-matrices and reconstruct a low-rank embedding. Hamburgers with different MDs can perform favorably against the popular global context module self-attention when carefully coping with gradients back-propagated through MDs. Comprehensive experiments are conducted in the vision tasks where it is crucial to learn the global context, including semantic segmentation and image generation, demonstrating significant improvements over self-attention and its variants.
In Syntology View this paper on Syntology: its repositories, every harvested function with whether it ran, its licence and the call to fetch it.
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="2109.04553")
Code
Syntology Ran 3 of 3 code samples harvested from 1 repository linked to this paper; 0 have no recorded run. Of those that ran: 3 ran with no contract checked.
By repository: community (archive-listed): 3 samples from 1 repository, 3 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.
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
3 samples harvested; 3 ran; 0 honoured the contract we drafted; 0 have no recorded run. Read from Syntology's graph 2026-09-25; that is when this build read the record, not when the samples ran.
Licence: 3 of the 3 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 plumprc/MTS-Mixers. “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.
cbc852fb32a663c7 · report
27b050d7deced77c · report
e5482e43844f6b74 · report
Tasks
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Conditional Image Generation | ImageNet 128x128 | HamGAN | FID | 14.80 | #18 of 22 | Archive leaderboard | report |
| Conditional Image Generation | ImageNet 128x128 | HamGAN | Inception score | 58.75 | #18 of 22 | Archive leaderboard | report |
| Semantic Segmentation | ADE20K | Light-Ham (VAN-Huge) | GFLOPs (512 x 512) | 71.8 | #95 of 235 | Archive leaderboard | report |
| Semantic Segmentation | ADE20K | Light-Ham (VAN-Huge) | Params (M) | 61.1 | #95 of 235 | Archive leaderboard | report |
| Semantic Segmentation | ADE20K | Light-Ham (VAN-Huge) | Validation mIoU | 51.5 | #95 of 235 | Archive leaderboard | report |
| Semantic Segmentation | ADE20K | Light-Ham (VAN-Large) | GFLOPs (512 x 512) | 55.0 | #103 of 235 | Archive leaderboard | report |
| Semantic Segmentation | ADE20K | Light-Ham (VAN-Large) | Params (M) | 45.6 | #103 of 235 | Archive leaderboard | report |
| Semantic Segmentation | ADE20K | Light-Ham (VAN-Large) | Validation mIoU | 51.0 | #103 of 235 | Archive leaderboard | report |
| Semantic Segmentation | ADE20K | Light-Ham (VAN-Base) | GFLOPs (512 x 512) | 34.4 | #130 of 235 | Archive leaderboard | report |
| Semantic Segmentation | ADE20K | Light-Ham (VAN-Base) | Params (M) | 27.4 | #130 of 235 | Archive leaderboard | report |
| Semantic Segmentation | ADE20K | Light-Ham (VAN-Base) | Validation mIoU | 49.6 | #130 of 235 | Archive leaderboard | report |
| Semantic Segmentation | ADE20K | HamNet (ResNet-101) | Validation mIoU | 46.8 | #170 of 235 | Archive leaderboard | report |
| Semantic Segmentation | ADE20K | Light-Ham (VAN-Small, D=256) | GFLOPs (512 x 512) | 15.8 | #195 of 235 | Archive leaderboard | report |
| Semantic Segmentation | ADE20K | Light-Ham (VAN-Small, D=256) | Params (M) | 13.8 | #195 of 235 | Archive leaderboard | report |
| Semantic Segmentation | ADE20K | Light-Ham (VAN-Small, D=256) | Validation mIoU | 45.2 | #195 of 235 | Archive leaderboard | report |
| Semantic Segmentation | ADE20K val | Light-Ham (VAN-Huge, 61M, IN-1k, MS) | mIoU | 51.5 | #45 of 95 | Archive leaderboard | report |
| Semantic Segmentation | ADE20K val | Light-Ham (VAN-Large, 46M, IN-1k, MS) | mIoU | 51.0 | #48 of 95 | Archive leaderboard | report |
| Semantic Segmentation | ADE20K val | Light-Ham (VAN-Base, 27M, IN-1k, MS) | mIoU | 49.6 | #58 of 95 | Archive leaderboard | report |
| Semantic Segmentation | PASCAL Context | HamNet (ResNet-101) | mIoU | 55.2 | #31 of 66 | Archive leaderboard | report |
| Semantic Segmentation | PASCAL VOC 2012 test | HamNet w/o COCO (ResNet-101) | Mean IoU | 85.9% | #8 of 51 | Archive leaderboard | report |
Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.
Methods
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