English
Related papers

Related papers: Towards Infinite Length Extrapolation: A Unified A…

200 papers

Transformer-based language models rely on positional encoding (PE) to handle token order and support context length extrapolation. However, existing PE methods lack theoretical clarity and rely on limited evaluation metrics to substantiate…

Computation and Language · Computer Science 2026-05-11 Arthur S. Bianchessi , Yasmin C. Aguirre , Rodrigo C. Barros , Lucas S. Kupssinskü

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…

Computation and Language · Computer Science 2025-10-24 Bowen Yang , Bharat Venkitesh , Dwarak Talupuru , Hangyu Lin , David Cairuz , Phil Blunsom , Acyr Locatelli

Since self-attention layers in Transformers are permutation invariant by design, positional encodings must be explicitly incorporated to enable spatial understanding. However, fixed-size lookup tables used in traditional learnable position…

Machine Learning · Computer Science 2025-06-18 Huayang Li , Yahui Liu , Hongyu Sun , Deng Cai , Leyang Cui , Wei Bi , Peilin Zhao , Taro Watanabe

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…

Computation and Language · Computer Science 2025-12-09 Xiaoran Liu , Yuerong Song , Zhigeng Liu , Zengfeng Huang , Qipeng Guo , Zhaoxiang Liu , Shiguo Lian , Ziwei He , Xipeng Qiu

We present Position Interpolation (PI) that extends the context window sizes of RoPE-based pretrained LLMs such as LLaMA models to up to 32768 with minimal fine-tuning (within 1000 steps), while demonstrating strong empirical results on…

Computation and Language · Computer Science 2023-06-29 Shouyuan Chen , Sherman Wong , Liangjian Chen , Yuandong Tian

Recently, Large language models (LLMs) have revolutionized Natural Language Processing (NLP). Pretrained LLMs, due to limited training context size, struggle with handling long token sequences, limiting their performance on various…

Computation and Language · Computer Science 2024-12-11 Haoran Lian , Junmin Chen , Wei Huang , Yizhe Xiong , Wenping Hu , Guiguang Ding , Hui Chen , Jianwei Niu , Zijia Lin , Fuzheng Zhang , Di Zhang

Transformers often struggle to generalize to longer sequences than those seen during training, a limitation known as length extrapolation. Most existing Relative Positional Encoding (RPE) methods attempt to address this by introducing…

Computation and Language · Computer Science 2025-09-23 Ali Veisi , Hamidreza Amirzadeh , Amir Mansourian

Understanding the internal representations of large language models (LLMs) remains a central challenge for interpretability research. Sparse autoencoders (SAEs) offer a promising solution by decomposing activations into interpretable…

Machine Learning · Computer Science 2025-10-10 Yifei Yao , Mengnan Du

Large Language Models (LLMs) have so far impressed the world, with unprecedented capabilities that emerge in models at large scales. On the vision side, transformer models (i.e., ViT) are following the same trend, achieving the best…

Computer Vision and Pattern Recognition · Computer Science 2023-10-30 Mustafa Shukor , Corentin Dancette , Matthieu Cord

The long-context capability of the Large Language Models (LLM) has made significant breakthroughs, but the maximum supported context length in length extrapolation remains a critical bottleneck limiting their practical applications. The…

Computation and Language · Computer Science 2025-03-20 Xiaoran Liu , Ruixiao Li , Qipeng Guo , Zhigeng Liu , Yuerong Song , Kai Lv , Hang Yan , Linlin Li , Qun Liu , Xipeng Qiu

Despite the remarkable capabilities of Multimodal Large Language Models (MLLMs), they still suffer from visual fading in long-context scenarios. Specifically, the attention to visual tokens diminishes as the text sequence lengthens, leading…

Computer Vision and Pattern Recognition · Computer Science 2026-03-12 Lin Chen , Bolin Ni , Qi Yang , Zili Wang , Kun Ding , Ying Wang , Houwen Peng , Shiming Xiang

Today's large language models (LLMs) typically train on short text segments (e.g., <4K tokens) due to the quadratic complexity of their Transformer architectures. As a result, their performance suffers drastically on inputs longer than…

Computation and Language · Computer Science 2024-06-26 Chi Han , Qifan Wang , Hao Peng , Wenhan Xiong , Yu Chen , Heng Ji , Sinong Wang

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…

Computation and Language · Computer Science 2026-02-06 Haoran Li , Sucheng Ren , Alan Yuille , Feng Wang

In Transformer-based architectures, the attention mechanism is inherently permutation-invariant with respect to the input sequence's tokens. To impose sequential order, token positions are typically encoded using a scheme with either fixed…

Machine Learning · Computer Science 2023-10-31 Giorgio Angelotti

Large Language Models (LLMs) are increasingly used in production systems, powering applications such as chatbots, summarization, and question answering. Despite their success, controlling the length of their response remains a significant…

Computation and Language · Computer Science 2025-05-12 Bradley Butcher , Michael O'Keefe , James Titchener

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…

Software Engineering · Computer Science 2024-08-12 Kechi Zhang , Ge Li , Huangzhao Zhang , Zhi Jin

Handling long-context sequences efficiently remains a significant challenge in large language models (LLMs). Existing methods for token selection in sequence extrapolation either employ a permanent eviction strategy or select tokens by…

Computation and Language · Computer Science 2025-02-21 Haoyu Wang , Tong Teng , Tianyu Guo , An Xiao , Duyu Tang , Hanting Chen , Yunhe Wang

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…

Computation and Language · Computer Science 2024-02-22 Dawei Zhu , Nan Yang , Liang Wang , Yifan Song , Wenhao Wu , Furu Wei , Sujian Li

Large Language Models (LLMs) often experience performance degradation during long-running interactions due to increasing context length, memory saturation, and computational overhead. This paper presents an adaptive context compression…

Computer Vision and Pattern Recognition · Computer Science 2026-04-01 Payal Fofadiya , Sunil Tiwari

Generalizing to longer sentences is important for recent Transformer-based language models. Besides algorithms manipulating explicit position features, the success of Transformers without position encodings (NoPE) provides a new way to…

Computation and Language · Computer Science 2024-05-29 Jie Wang , Tao Ji , Yuanbin Wu , Hang Yan , Tao Gui , Qi Zhang , Xuanjing Huang , Xiaoling Wang