English
Related papers

Related papers: Extending Context Window of Large Language Models …

200 papers

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

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

Information retrieval in Large Language Models (LLMs) is increasingly recognized as intertwined with generation capabilities rather than mere lookup. While longer contexts are often assumed to improve retrieval, the effects of intra-context…

Computation and Language · Computer Science 2025-08-01 Chupei Wang , Jiaqiu Vince Sun

Transformer-based Large Language Models (LLMs) struggle with inputs exceeding their training context window due to positional out-of-distribution (O.O.D.) issues that disrupt attention. Existing solutions, including fine-tuning and…

Computation and Language · Computer Science 2025-06-02 Yan Li , Tianyi Zhang , Zechuan Li , Soyeon Caren Han

Effectively training language models on long inputs poses many technical challenges. As a cost consideration, languages models are pretrained on a fixed sequence length before being adapted to longer sequences. We explore various methods…

Computation and Language · Computer Science 2024-06-21 Petros Karypis , Julian McAuley , George Karypis

Using special tokens (e.g., gist, memory, or compressed tokens) to compress context information is a common practice for large language models (LLMs). However, existing approaches often neglect that position encodings inherently induce…

Computation and Language · Computer Science 2025-09-29 Runsong Zhao , Xin Liu , Xinyu Liu , Pengcheng Huang , Chunyang Xiao , Tong Xiao , Jingbo Zhu

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.…

Computation and Language · Computer Science 2025-02-18 Ting-Rui Chiang , Dani Yogatama

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

Extending the context window of large language models (LLMs) is getting popular recently, while the solution of augmenting LLMs with retrieval has existed for years. The natural questions are: i) Retrieval-augmentation versus long context…

Computation and Language · Computer Science 2024-01-24 Peng Xu , Wei Ping , Xianchao Wu , Lawrence McAfee , Chen Zhu , Zihan Liu , Sandeep Subramanian , Evelina Bakhturina , Mohammad Shoeybi , Bryan Catanzaro

While large vision-language models (VLMs) demonstrate strong long-context understanding, their prevalent small branches fail on linguistics-photography alignment for a limited window size. We discover that knowledge distillation improves…

Computer Vision and Pattern Recognition · Computer Science 2026-01-06 Haoyi Zhou , Shuo Li , Tianyu Chen , Qi Song , Chonghan Gao , Jianxin Li

Extending the context window of language models typically requires expensive long-context pre-training, posing significant challenges for both training efficiency and data collection. In this paper, we present evidence that long-context…

Computation and Language · Computer Science 2026-04-08 Patrick Huber , Ernie Chang , Chinnadhurai Sankar , Rylan Conway , Igor Fedorov , Md Rifat Arefin , Adithya Sagar

This paper addresses the challenge of train-short-test-long (TSTL) scenarios in Large Language Models (LLMs) equipped with Rotary Position Embedding (RoPE), where models pre-trained on shorter sequences face difficulty with…

Computation and Language · Computer Science 2024-09-05 Suyuchen Wang , Ivan Kobyzev , Peng Lu , Mehdi Rezagholizadeh , Bang Liu

Typically, training LLMs with long context sizes is computationally expensive, requiring extensive training hours and GPU resources. Existing long-context extension methods usually need additional training procedures to support…

Computation and Language · Computer Science 2024-02-23 Jiaheng Liu , Zhiqi Bai , Yuanxing Zhang , Chenchen Zhang , Yu Zhang , Ge Zhang , Jiakai Wang , Haoran Que , Yukang Chen , Wenbo Su , Tiezheng Ge , Jie Fu , Wenhu Chen , Bo Zheng

The context window of large language models has been extended to 128k tokens or more. However, language models still suffer from position bias and have difficulty in accessing and using the middle part of the context due to the lack of…

Computation and Language · Computer Science 2024-06-26 Meiru Zhang , Zaiqiao Meng , Nigel Collier

We study the post-training of large language models (LLMs) with human preference data. Recently, direct preference optimization and its variants have shown considerable promise in aligning language models, eliminating the need for reward…

Machine Learning · Computer Science 2025-05-14 Teng Xiao , Zhen Ge , Sujay Sanghavi , Tian Wang , Julian Katz-Samuels , Marc Versage , Qingjun Cui , Trishul Chilimbi

We present LongQLoRA, an efficient and effective method to extend context length of large language models with less training resources. LongQLoRA combines the advantages of Position Interpolation, QLoRA and Shift Short Attention of…

Computation and Language · Computer Science 2023-11-10 Jianxin Yang

Large Language Models (LLMs) often struggle to use information across long inputs effectively. Prior work has identified positional biases, such as the Lost in the Middle (LiM) effect, where models perform better when information appears at…

Computation and Language · Computer Science 2025-08-12 Blerta Veseli , Julian Chibane , Mariya Toneva , Alexander Koller

Transformer-based Large Language Models (LLMs) often impose limitations on the length of the text input to ensure the generation of fluent and relevant responses. This constraint restricts their applicability in scenarios involving long…

Computation and Language · Computer Science 2023-12-18 Weizhi Fei , Xueyan Niu , Pingyi Zhou , Lu Hou , Bo Bai , Lei Deng , Wei Han

The advent of Large Language Models (LLMs) represents a notable breakthrough in Natural Language Processing (NLP), contributing to substantial progress in both text comprehension and generation. However, amidst these advancements, it is…

Computation and Language · Computer Science 2024-01-17 Saurav Pawar , S. M Towhidul Islam Tonmoy , S M Mehedi Zaman , Vinija Jain , Aman Chadha , Amitava Das

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…

Computation and Language · Computer Science 2026-05-18 Yufeng Du , Phillip Harris , Minyang Tian , Eliu A Huerta , Srikanth Ronanki , Subendhu Rongali , Aram Galstyan , Hao Peng