Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity
Work-level record. Its arXiv versions and any conference or journal version are listed as sources, each with its own date. A citation to the work does not identify which version the citing author read.
- Issued by
- No issuing record yet. A missing record is a gap in coverage.
- Authors credited
- Barret Zoph · Liam Fedus · Noam Shazeer
- Dates
- Published 2021-01-11 authorship evidence authorship evidence authorship evidence
- Exact source
- Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity arXiv · published 2021-01-11 · arXiv abstract page, retrieved for the candidate pack on 2026-10-28 (SHA-256 3a6496323b841d8fe79e5026614930e69ba594a480cc0d72c70c47a2869f5c6a). Submission history as arXiv records it (UTC): v1 11 January 2021; v2 30 April 2022; v3 16 June 2022. The page serves the latest version; each version is a separate expression, and only the one a record names was read.
- Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity arXiv · published 2021-01-11 · arXiv:2101.03961v1, submitted 11 January 2021 (UTC). Retrieved for the candidate pack on 2026-10-28 (SHA-256 0dba4abc34422f49a5363f63c90d2103e4a982c921c597b8c3560cc34fd80d9c). The author list is the byline as printed on the first page.
Appears in
No collection holds a record about this document yet.
Versions
No other versions of this document are recorded. The evidence applies to the source version listed above. Other versions, such as a preprint, working paper or journal publication, are not identified or substituted.
Issue and authorship records
- Author: Barret ZophBarret Zoph is a credited author of Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity.2021-01-11 · publication · Stated in source · Supported Inspect claim & evidence ↗Barret Zoph: person page →
- Author: Liam FedusWilliam Fedus is a credited author of Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity.2021-01-11 · publication · Stated in source · Supported Inspect claim & evidence ↗Liam Fedus: person page →
- Author: Noam ShazeerNoam Shazeer is a credited author of Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity.2021-01-11 · publication · Stated in source · Supported Inspect claim & evidence ↗Noam Shazeer: person page →
Documentary relations
Each relation is listed separately. Citing, crediting and reporting the use of a method are different acts, and none of them shows agreement or influence by itself.
Cited by · 7
Recorded documents that explicitly reference this one. A citation to a work does not identify which version was read.
- ERNIE 3.0: Large-scale Knowledge Enhanced Pre-training for Language Understanding and GenerationERNIE 3.0: Large-scale Knowledge Enhanced Pre-training for Language Understanding and Generation cites Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity.2021-07-05 · publication · Stated in source · Supported Inspect claim & evidence ↗ERNIE 3.0: Large-scale Knowledge Enhanced Pre-training for Language Understanding and Generation: document page →
- ERNIE 3.0 Titan: Exploring Larger-scale Knowledge Enhanced Pre-training for Language Understanding and GenerationERNIE 3.0 Titan: Exploring Larger-scale Knowledge Enhanced Pre-training for Language Understanding and Generation cites Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity.2021-12-23 · publication · Stated in source · Supported Inspect claim & evidence ↗ERNIE 3.0 Titan: Exploring Larger-scale Knowledge Enhanced Pre-training for Language Understanding and Generation: document page →
- Training language models to follow instructions with human feedbackTraining language models to follow instructions with human feedback cites Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity.2022-03-04 · publication · Stated in source · Supported Inspect claim & evidence ↗Training language models to follow instructions with human feedback: document page →
- Training Compute-Optimal Large Language ModelsTraining Compute-Optimal Large Language Models cites Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity.2022-03-29 · publication · Stated in source · Supported Inspect claim & evidence ↗Training Compute-Optimal Large Language Models: document page →
- Qwen Technical ReportQwen Technical Report cites Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity.2023-09-28 · publication · Stated in source · Supported Inspect claim & evidence ↗Qwen Technical Report: document page →
- DeepSeek-V3 Technical ReportDeepSeek-V3 Technical Report cites Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity.2024-12-27 · publication · Stated in source · Supported Inspect claim & evidence ↗DeepSeek-V3 Technical Report: document page →
- Kimi K3: Open Frontier IntelligenceKimi K3: Open Frontier Intelligence cites Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity.2026-07-28 · publication · Stated in source · Supported Inspect claim & evidence ↗Kimi K3: Open Frontier Intelligence: document page →
Not recorded for this document: cites · explicitly credits · reports using a method from · method reported as used by · applies parts of · parts applied by · discusses · reports evaluation of. A missing record does not show that the relation does not exist.
What the document says it adds
The document’s own description of its contribution, quoted from the version each row names. A claim to be the first is the document’s claim; Selekoa does not assess it.
- In its own wordsThe paper presents the Switch Transformer as its response to the complexity, communication costs and training instability it attributes to Mixture of Experts models.“We address these with the Switch Transformer.” arXiv v1, PDF p. 1, abstract2021-01-11 · publication · Stated in source · Supported Inspect claim & evidence ↗
- In its own wordsThe authors state that they simplify the Mixture of Experts routing algorithm and design models with reduced communication and computational costs.“We simplify the MoE routing algorithm and design intuitive improved models with reduced communication and computational costs.” arXiv v1, PDF p. 1, abstract2021-01-11 · publication · Stated in source · Supported Inspect claim & evidence ↗