中文
相关论文

相关论文: Understanding the RoPE Extensions of Long-Context …

200 篇论文

Large Language Models (LLMs) are known to have limited extrapolation ability beyond their pre-trained context window, constraining their application in downstream tasks with lengthy inputs. Recent studies have sought to extend LLMs' context…

计算与语言 · 计算机科学 2024-01-17 Yikai Zhang , Junlong Li , Pengfei Liu

The extrapolation capability of Large Language Models (LLMs) based on Rotary Position Embedding is currently a topic of considerable interest. The mainstream approach to addressing extrapolation with LLMs involves modifying RoPE by…

计算与语言 · 计算机科学 2024-03-14 Xiaoran Liu , Hang Yan , Shuo Zhang , Chenxin An , Xipeng Qiu , Dahua Lin

Built upon the Transformer, large language models (LLMs) have captured worldwide attention due to their remarkable abilities. Nevertheless, all Transformer-based models including LLMs suffer from a preset length limit and can hardly…

计算与语言 · 计算机科学 2024-10-08 Liang Zhao , Xiachong Feng , Xiaocheng Feng , Weihong Zhong , Dongliang Xu , Qing Yang , Hongtao Liu , Bing Qin , Ting Liu

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…

计算与语言 · 计算机科学 2024-10-22 Xiangyu Hong , Che Jiang , Biqing Qi , Fandong Meng , Mo Yu , Bowen Zhou , Jie Zhou

Position embedding is a core component of current Large Language Models (LLMs). Rotary position embedding (RoPE), a technique that encodes the position information with a rotation matrix, has been the de facto choice for position embedding…

计算与语言 · 计算机科学 2024-05-24 Xin Men , Mingyu Xu , Bingning Wang , Qingyu Zhang , Hongyu Lin , Xianpei Han , Weipeng Chen

Addressing the limitation of context length in large language models for code-related tasks is the primary focus of this paper. Existing LLMs are constrained by their pre-trained context lengths, leading to performance issues in handling…

软件工程 · 计算机科学 2024-08-12 Kechi Zhang , Ge Li , Huangzhao Zhang , Zhi Jin

Rotary Position Embeddings (RoPE) have become a standard for encoding sequence order in Large Language Models (LLMs) by applying rotations to query and key vectors in the complex plane. Standard implementations, however, utilize only the…

计算与语言 · 计算机科学 2025-12-09 Xiaoran Liu , Yuerong Song , Zhigeng Liu , Zengfeng Huang , Qipeng Guo , Zhaoxiang Liu , Shiguo Lian , Ziwei He , Xipeng Qiu

The rapid advancement of large language models (LLMs) has led to a significant increase in automated tools in the software engineering, capable of performing various code-related tasks such as code generation, completion, and translation.…

软件工程 · 计算机科学 2026-02-26 Madhusudan Ghosh , Rishabh Gupta

Many positional encodings (PEs) are designed to exhibit long-term decay, based on an entrenched and long-standing inductive opinion: tokens farther away from the current position carry less relevant information. We argue that long-term…

计算与语言 · 计算机科学 2024-12-06 Yuhan Chen , Ang Lv , Jian Luan , Bin Wang , Wei Liu

Long-context large language models (LLMs) have achieved remarkable advancements, driven by techniques like Rotary Position Embedding (RoPE) (Su et al., 2023) and its extensions (Chen et al., 2023; Liu et al., 2024c; Peng et al., 2023). By…

计算与语言 · 计算机科学 2025-10-24 Bowen Yang , Bharat Venkitesh , Dwarak Talupuru , Hangyu Lin , David Cairuz , Phil Blunsom , Acyr Locatelli

Extending context window sizes allows large language models (LLMs) to process longer sequences and handle more complex tasks. Rotary Positional Embedding (RoPE) has become the de facto standard due to its relative positional encoding…

计算与语言 · 计算机科学 2024-11-27 Haonan Wang , Qian Liu , Chao Du , Tongyao Zhu , Cunxiao Du , Kenji Kawaguchi , Tianyu Pang

We identify intrinsic limitations of Rotary Positional Embeddings (RoPE) in Transformer-based long-context language models. Our theoretical analysis abstracts away from the specific content of the context and depends only on its length. We…

计算与语言 · 计算机科学 2026-05-18 Yufeng Du , Phillip Harris , Minyang Tian , Eliu A Huerta , Srikanth Ronanki , Subendhu Rongali , Aram Galstyan , Hao Peng

The Rotary Position Embedding (RoPE) is widely used in the attention heads of many large language models (LLM). It rotates dimensions in the query and the key vectors by different angles according to their positions in the input sequence.…

计算与语言 · 计算机科学 2025-02-18 Ting-Rui Chiang , Dani Yogatama

Rotary Positional Embedding (RoPE) is a key component of context scaling in Large Language Models (LLMs). While various methods have been proposed to adapt RoPE to longer contexts, their guiding principles generally fall into two…

计算与语言 · 计算机科学 2026-02-06 Haoran Li , Sucheng Ren , Alan Yuille , Feng Wang

Broad textual understanding and in-context learning require language models that utilize full document contexts. Due to the implementation challenges associated with directly training long-context models, many methods have been proposed for…

计算与语言 · 计算机科学 2024-09-24 Yi Lu , Jing Nathan Yan , Songlin Yang , Justin T. Chiu , Siyu Ren , Fei Yuan , Wenting Zhao , Zhiyong Wu , Alexander M. Rush

In the realm of large-scale language models, a significant challenge arises when extrapolating sequences beyond the maximum allowable length. This is because the model's position embedding mechanisms are limited to positions encountered…

计算与语言 · 计算机科学 2025-02-05 Yui Oka , Taku Hasegawa , Kyosuke Nishida , Kuniko Saito

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…

计算与语言 · 计算机科学 2024-03-26 Guanzheng Chen , Xin Li , Zaiqiao Meng , Shangsong Liang , Lidong Bing

Large language models (LLMs), although having revolutionized many fields, still suffer from the challenging extrapolation problem, where the inference ability of LLMs sharply declines beyond their max training lengths. In this work, we…

机器学习 · 计算机科学 2024-10-25 Xin Ma , Yang Liu , Jingjing Liu , Xiaoxu Ma

Embedding models play a pivot role in modern NLP applications such as IR and RAG. While the context limit of LLMs has been pushed beyond 1 million tokens, embedding models are still confined to a narrow context window not exceeding 8k…

计算与语言 · 计算机科学 2024-11-08 Dawei Zhu , Liang Wang , Nan Yang , Yifan Song , Wenhao Wu , Furu Wei , Sujian Li

Large Language Models (LLMs) are trained with a pre-defined context length, restricting their use in scenarios requiring long inputs. Previous efforts for adapting LLMs to a longer length usually requires fine-tuning with this target length…

计算与语言 · 计算机科学 2024-02-22 Dawei Zhu , Nan Yang , Liang Wang , Yifan Song , Wenhao Wu , Furu Wei , Sujian Li
‹ 上一页 1 2 3 10 下一页 ›