English
Related papers

Related papers: Frayed RoPE and Long Inputs: A Geometric Perspecti…

200 papers

Large language models (LLMs), even when specifically trained to process long input contexts, struggle to capture relevant information located in the middle of their input. This phenomenon has been known as the lost-in-the-middle problem. In…

In this paper, we demonstrate that an inherent waveform pattern in the attention allocation of large language models (LLMs) significantly affects their performance in tasks demanding a high degree of context awareness, such as utilizing…

Computation and Language · Computer Science 2024-06-05 Yuhan Chen , Ang Lv , Ting-En Lin , Changyu Chen , Yuchuan Wu , Fei Huang , Yongbin Li , Rui Yan

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

The goal of this paper is to strengthen the reasoning of Omnimodal Large Language Models (Omni-LLMs) at inference time, without additional training. These models jointly process video, audio, and text, and given the large number of tokens…

Computer Vision and Pattern Recognition · Computer Science 2026-05-08 Suho Yoo , Youngjoon Jang , Joon Son Chung

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…

Computation and Language · Computer Science 2024-10-08 Liang Zhao , Xiachong Feng , Xiaocheng Feng , Weihong Zhong , Dongliang Xu , Qing Yang , Hongtao Liu , Bing Qin , Ting Liu

Sparse Mixture of Experts (MoE) models are popular for training large language models due to their computational efficiency. However, the commonly used top-$k$ routing mechanism suffers from redundancy computation and memory costs due to…

Machine Learning · Computer Science 2024-02-22 Zhiyuan Zeng , Qipeng Guo , Zhaoye Fei , Zhangyue Yin , Yunhua Zhou , Linyang Li , Tianxiang Sun , Hang Yan , Dahua Lin , Xipeng Qiu

Vision-based pose estimation of articulated robots with unknown joint angles has applications in collaborative robotics and human-robot interaction tasks. Current frameworks use neural network encoders to extract image features and…

Robotics · Computer Science 2025-05-05 Raktim Gautam Goswami , Prashanth Krishnamurthy , Yann LeCun , Farshad Khorrami

Transformer-based models have become the dominant paradigm for neural combinatorial optimization (NCO) of vehicle routing problems (VRPs), yet the role of positional encoding (PE) in these architectures remains largely unexplored. Unlike…

Deep learning models, particularly Transformers, are often criticized as "black boxes" and lack interpretability. We propose Prism, a white-box attention-based architecture derived from the principles of Maximizing Coding Rate Reduction…

Machine Learning · Computer Science 2026-02-04 Dongchen Huang

Position embeddings, encoding the positional relationships among tokens in text sequences, make great contributions to modeling local context features in Transformer-based pre-trained language models. However, in Extractive Question…

Computation and Language · Computer Science 2023-11-21 Mingxu Tao , Yansong Feng , Dongyan Zhao

Transformer-based pre-trained models, such as BERT, have achieved remarkable results on machine reading comprehension. However, due to the constraint of encoding length (e.g., 512 WordPiece tokens), a long document is usually split into…

Computation and Language · Computer Science 2021-09-15 Jing Zhao , Junwei Bao , Yifan Wang , Yongwei Zhou , Youzheng Wu , Xiaodong He , Bowen Zhou

Large language models have shown remarkable performance across a wide range of language tasks, owing to their exceptional capabilities in context modeling. The most commonly used method of context modeling is full self-attention, as seen in…

Computation and Language · Computer Science 2025-06-26 Zhisong Zhang , Yan Wang , Xinting Huang , Tianqing Fang , Hongming Zhang , Chenlong Deng , Shuaiyi Li , Dong Yu

In Transformers, Position Embeddings (PEs) significantly influence Length Generalization (LG) performance, yet their fundamental role remains unclear. In this work, we investigate the limitations and capabilities of PEs in achieving LG. We…

Machine Learning · Computer Science 2025-10-07 Yang Chen , Yitao Liang , Zhouchen Lin

Vision-Language Models suffer severe KV cache pressure at inference, as a single image often encodes into thousands of tokens. Most existing methods exploit token sparsity through token pruning, but permanently discarding visual content…

Computer Vision and Pattern Recognition · Computer Science 2026-05-20 Beomseok Kang , Dongwon Jo , Jiwon Song , Donghwee Son , Jae-Joon Kim

Position encoding (PE), an essential part of self-attention networks (SANs), is used to preserve the word order information for natural language processing tasks, generating fixed position indices for input sequences. However, in…

Computation and Language · Computer Science 2020-11-24 Liang Ding , Longyue Wang , Dacheng Tao

In Transformer architectures, tokens\textemdash discrete units derived from raw data\textemdash are formed by segmenting inputs into fixed-length chunks. Each token is then mapped to an embedding, enabling parallel attention computations…

Machine Learning · Computer Science 2026-01-14 Zhenglun Kong , Yize Li , Fanhu Zeng , Lei Xin , Shvat Messica , Xue Lin , Pu Zhao , Manolis Kellis , Hao Tang , Marinka Zitnik

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

Lifelong user behavior sequences are crucial for capturing user interests and predicting user responses in modern recommendation systems. A two-stage paradigm is typically adopted to handle these long sequences: a subset of relevant…

Information Retrieval · Computer Science 2025-03-27 Ningya Feng , Junwei Pan , Jialong Wu , Baixu Chen , Ximei Wang , Qian Li , Xian Hu , Jie Jiang , Mingsheng Long

Large language models (LLMs) have revolutionized natural language processing, but their ability to process long sequences is fundamentally limited by the context window size during training. Existing length extrapolation methods often…

Artificial Intelligence · Computer Science 2026-01-13 Nitin Vetcha

Scaling the size of a model enhances its capabilities but significantly increases computation complexity. Mixture-of-Experts models (MoE) address the issue by allowing model size to scale up without substantially increasing training or…

Computation and Language · Computer Science 2024-08-30 Zhenpeng Su , Zijia Lin , Xue Bai , Xing Wu , Yizhe Xiong , Haoran Lian , Guangyuan Ma , Hui Chen , Guiguang Ding , Wei Zhou , Songlin Hu