Skip to main content
PaperLLMsv1.0

Generating Long Sequences with Sparse Transformers

by OpenAI · free · Last verified 2026-03-17

Introduces Sparse Transformers, which use factored sparse attention patterns to reduce attention complexity from O(n²) to O(n√n), enabling transformers to model sequences thousands of steps long. Applied to autoregressive density estimation of images, text, and audio.

https://arxiv.org/abs/1904.10509
B
BAbove Average
Adoption: BQuality: AFreshness: DCitations: BEngagement: F

Specifications

License
Open Access
Pricing
free
Capabilities
sparse-attention, long-context-modeling, density-estimation
Integrations
Use Cases
long-sequence-generation, image-generation, audio-generation
API Available
No
Tags
sparse-attention, long-context, transformers, generative-modeling, openai
Added
2026-03-17
Completeness
100%

Index Score

61.2
Adoption
68
Quality
85
Freshness
38
Citations
68
Engagement
0

Put AI to work for your business

Deploy this paper alongside autonomous AaaS agents that handle tasks end-to-end — no babysitting required.

Explore the full AI ecosystem on Agents as a Service