Machine learningML & AI Methods
What Algorithms can Transformers Learn? A Study in Length Generalization
The study suggests that Transformers can demonstrate strong length generalization on tasks that can be solved by a short RASP program applicable to all input lengths.
Featured in No. 23 on 25 Oct 2023 · 1 day after release · 218 citations today · published in International Conference on Learning Representations
- Released
- 24 Oct 2023
- First featured
- No. 23 · 25 Oct 2023
- Citations (Semantic Scholar)
- 218
- Influential citations
- 20
- Published in
- International Conference on Learning Representations
- Shares when featured
- 93
- Identifier
- arXiv:2310.16028
Citations and venue from Semantic Scholar (ODC-BY), refreshed weekly. Summary: Quant Letter (CC BY 4.0).