arXiv:2607. 18759v1 Announce Type: new Abstract: Transformers with relative positional encodings often extrapolate to sequences longer than those seen during training, whereas transformers with learned absolute encodings typically do not.
By Subham Singh, Ashutosh Mishra, Subha Raut
arXiv:2608. 10251v1 Announce Type: cross Abstract: A transformer's answer lives on one axis: the direction its unembedding reads.
By Mark Oskin
arXiv:2511. 17388v3 Announce Type: replace-cross Abstract: Position information is essential for language modeling.
By Sajad Movahedi, Timur Carstensen, Arshia Afzal, Frank Hutter, Antonio Orvieto, Volkan Cevher
The projection of queries and keys are central to the attention mechanism in Transformer architectures. While they are mathematically symmetric, they play different roles in attention mechanisms.
arXiv:2607. 10134v1 Announce Type: new Abstract: Rotary Positional Encodings (RoPE) are currently the most popular positional encodings used in modern language models.
By Petros Karypis, Sean O'Brien, Shreyas Kadekodi, Rui Zhu, Julian McAuley
The paper proposes a content‑based addressing scheme for long‑context models that replaces the growing token counter in Rotary Position Embedding (RoPE) with unit‑level addresses derived from the content of each unit. By dividing the token stream into units, the method preserves local RoPE behavior while allowing new units to be addressed via learned content maps, avoiding positional mismatches when extending context length. Experiments on character‑level Tiny Shakespeare show that a model trained on 256‑character contexts achieves lower perplexity at 4096 characters using this scheme, and a second diagnostic demonstrates retrieval of multiple serialized facts.
By Mahesh Godavarti