English
Related papers

Related papers: ExPe: Exact Positional Encodings for Generative Tr…

200 papers

Relative positional encoding is widely used in vanilla and linear transformers to represent positional information. However, existing encoding methods of a vanilla transformer are not always directly applicable to a linear transformer,…

Computation and Language · Computer Science 2023-07-19 Zhen Qin , Weixuan Sun , Kaiyue Lu , Hui Deng , Dongxu Li , Xiaodong Han , Yuchao Dai , Lingpeng Kong , Yiran Zhong

Rotary Position Embedding (RoPE) performs remarkably on language models, especially for length extrapolation of Transformers. However, the impacts of RoPE on computer vision domains have been underexplored, even though RoPE appears capable…

Computer Vision and Pattern Recognition · Computer Science 2024-07-17 Byeongho Heo , Song Park , Dongyoon Han , Sangdoo Yun

Despite the success of Transformers on language understanding, code generation, and logical reasoning, they still fail to generalize over length on basic arithmetic tasks such as addition and multiplication. A major reason behind this…

Machine Learning · Computer Science 2024-06-05 Mahdi Sabbaghi , George Pappas , Hamed Hassani , Surbhi Goel

Length extrapolation algorithms based on Rotary position embedding (RoPE) have shown promising results in extending the context length of language models. However, understanding how position embedding can capture longer-range contextual…

Computation and Language · Computer Science 2024-10-22 Xiangyu Hong , Che Jiang , Biqing Qi , Fandong Meng , Mo Yu , Bowen Zhou , Jie Zhou

In transformer architectures, position encoding primarily provides a sense of sequence for input tokens. While the original transformer paper's method has shown satisfactory results in general language processing tasks, there have been new…

Computation and Language · Computer Science 2024-03-26 Eren Unlu

In this work, we leverage the intrinsic segmentation of language sequences and design a new positional encoding method called Bilevel Positional Encoding (BiPE). For each position, our BiPE blends an intra-segment encoding and an…

Machine Learning · Computer Science 2024-06-18 Zhenyu He , Guhao Feng , Shengjie Luo , Kai Yang , Liwei Wang , Jingjing Xu , Zhi Zhang , Hongxia Yang , Di He

Transformer-based Large Language Models (LLMs) are pioneering advances in many natural language processing tasks, however, their exceptional capabilities are restricted within the preset context window of Transformer. Position Embedding…

Computation and Language · Computer Science 2024-03-26 Guanzheng Chen , Xin Li , Zaiqiao Meng , Shangsong Liang , Lidong Bing

Standard Vision Transformers flatten 2D images into 1D sequences, disrupting the natural spatial topology. While Rotary Positional Embedding (RoPE) excels in 1D, it inherits this limitation, often treating spatially distant patches (e.g.,…

Computer Vision and Pattern Recognition · Computer Science 2025-12-05 Yupu Yao , Bowen Yang

This paper studies how Transformer models with Rotary Position Embeddings (RoPE) develop emergent, wavelet-like properties that compensate for the positional encoding's theoretical limitations. Through an analysis spanning model scales,…

Machine Learning · Computer Science 2025-06-06 Valeria Ruscio , Umberto Nanni , Fabrizio Silvestri

Embedding techniques have become essential components of large databases in the deep learning era. By encoding discrete entities, such as words, items, or graph nodes, into continuous vector spaces, embeddings facilitate more efficient…

Information Retrieval · Computer Science 2024-10-18 Shiwei Li , Zhuoqi Hu , Xing Tang , Haozhao Wang , Shijie Xu , Weihong Luo , Yuhua Li , Xiuqiang He , Ruixuan Li

Transformer architectures rely on position encodings to model the spatial structure of input data. Rotary Position Encoding (RoPE) is a widely used method in language models that encodes relative positions through fixed, block-diagonal,…

Computer Vision and Pattern Recognition · Computer Science 2025-08-19 Sophie Ostmeier , Brian Axelrod , Maya Varma , Michael E. Moseley , Akshay Chaudhari , Curtis Langlotz

Applying Transformers to irregular time-series typically requires specializations to their baseline architecture, which can result in additional computational overhead and increased method complexity. We present the Rotary Masked…

Machine Learning · Computer Science 2026-05-13 Uros Zivanovic , Serafina Di Gioia , Andre Scaffidi , Martín de los Rios , Gabriella Contardo , Roberto Trotta

Transformers with causal attention can solve tasks that require positional information without using positional encodings. In this work, we propose and investigate a new hypothesis about how positional information can be stored without…

Computation and Language · Computer Science 2025-01-03 Chunsheng Zuo , Pavel Guerzhoy , Michael Guerzhoy

Rotary Position Embedding (RoPE) is the de facto positional encoding in large language models due to its ability to encode relative positions and support length extrapolation. When adapted to vision transformers, the standard axial…

Computer Vision and Pattern Recognition · Computer Science 2026-02-04 Haoyu Liu , Sucheng Ren , Tingyu Zhu , Peng Wang , Cihang Xie , Alan Yuille , Zeyu Zheng , Feng Wang

Shape completion, a crucial task in 3D computer vision, involves predicting and filling the missing regions of scanned or partially observed objects. Current methods expect known pose or canonical coordinates and do not perform well under…

Computer Vision and Pattern Recognition · Computer Science 2024-12-03 Burak Bekci , Nassir Navab , Federico Tombari , Mahdi Saleh

Enabling LLMs to handle lengthy context is currently a research hotspot. Most LLMs are built upon rotary position embedding (RoPE), a popular position encoding method. Therefore, a prominent path is to extrapolate the RoPE trained on…

Computation and Language · Computer Science 2024-12-13 Meizhi Zhong , Chen Zhang , Yikun Lei , Xikai Liu , Yan Gao , Yao Hu , Kehai Chen , Min Zhang

We propose a new positional encoding method for a neural network architecture called the Transformer. Unlike the standard sinusoidal positional encoding, our approach is based on solid mathematical grounds and has a guarantee of not losing…

Machine Learning · Computer Science 2024-05-17 Tsuyoshi Idé , Jokin Labaien , Pin-Yu Chen

Positional Encodings (PEs) are used to inject word-order information into transformer-based language models. While they can significantly enhance the quality of sentence representations, their specific contribution to language models is not…

Computation and Language · Computer Science 2023-10-20 Lihu Chen , Gaël Varoquaux , Fabian M. Suchanek

Transformer models, which leverage architectural improvements like self-attention, perform remarkably well on Natural Language Processing (NLP) tasks. The self-attention mechanism is position agnostic. In order to capture positional…

Computation and Language · Computer Science 2021-09-28 Zhiheng Huang , Davis Liang , Peng Xu , Bing Xiang

Positional encoding is essential for large language models (LLMs) to represent sequence order, yet recent studies show that Rotary Position Embedding (RoPE) can induce massive activation. We investigate the source of these instabilities via…

Computation and Language · Computer Science 2026-01-07 Jing Xiong , Liyang Fan , Hui Shen , Zunhai Su , Min Yang , Lingpeng Kong , Ngai Wong