arXiv Machine Learning By Qian Li, Xinyu Mao, Shang-Hua Teng

Rethinking the Role of Positional Encoding: Sliding-Window Transformers without PE Remain Turing Complete

Read the original on arXiv Machine Learning →

arXiv:2606. 01532v1 Announce Type: new Abstract: Positional encoding (PE) is widely viewed as necessary for transformers to process ordered sequences: without them, the next-token map appears permutation-invariant in its context tokens.

Summary generated by The Flow from the publisher's feed. The full article lives at arXiv Machine Learning.

Hugging Face Trending Papers
Jun 1

Rethinking the Role of Positional Encoding: Sliding-Window Transformers without PE Remain Turing Complete

Positional encoding (PE) is widely viewed as necessary for transformers to process ordered sequences: without them, the next-token map appears permutation-invariant in its context tokens. This intuition underlies all prior universality results, which rely on positional information to prove that transformers with chain-of-thought can perform arbitrary computation, i.