AI Paper Atlas

Research map / Video / Video generation

SLA: Beyond Sparsity in Diffusion Transformers via Fine-Tunable Sparse–Linear Attention

Jintao Zhang, Haoxu Wang, Kai Jiang, Shuo Yang, Kaiwen Zheng, Haocheng Xi, Ziteng Wang, Hongzhou Zhu, Min Zhao, Ion Stoica, Joseph E. Gonzalez, Jianfei Chen and 1 more

ICLR 2026 · Video › Video generation · approach: Efficient video diffusion models

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
fuse linear attention for marginal blocks into sparse attention (SLA vs Sparse Only) (ablation)SLA · Ablation (Wan2.1-1.3B)VA: 64.00 -> 76.96Improved
add linear attention with sparse attention via learned fusion vs directly summing (L+S) (ablation)SLA · Ablation (Wan2.1-1.3B)VA: 29.65 -> 76.96Improved
use linear attention only (ablation)Wan2.1 · video genVA: 76.78 (full) -> 0.042Hurt
use softmax instead of elu+1 as linear attention activation (ablation)SLA · Ablation (Wan2.1-1.3B)VA: 75.50 -> 76.96Improved
use softmax instead of hedgehog activation (ablation)SLA · Ablation (Wan2.1-1.3B)VA: 74.59 -> 76.96Improved
reduce k_h from 20% to 5% (ablation)SLA · Ablation (Wan2.1-1.3B)VA: 75.81 -> 76.96 with FLOPs 10.65T -> 2.73TNo clear effect
raise sparsity to 95% with SLA vs baselines at 85-89%SLA vs VSA/VMoBa/Sparge · video genVA: VSA 55.37, VMoBa 32.33 vs SLA 76.96Improved
fine-tune with SLA vs full attentionWan2.1-1.3B · video genVA: 76.78 -> 76.96No clear effect
apply SLA to image generationLightningDiT · ImageNet 512FID: 31.87 -> 31.49Improved
use SLA kernel vs FlashAttention2attention kernel · RTX5090 Wan2.1kernel speed: 13.7x forward, 6.8x backwardImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy