English
Related papers

Related papers: Token-Picker: Accelerating Attention in Text Gener…

200 papers

Linear attention offers a computationally efficient yet expressive alternative to softmax attention. However, recent empirical results indicate that the hidden state of trained linear attention models often exhibits a low-rank structure,…

Machine Learning · Computer Science 2026-02-13 Philipp Nazari , T. Konstantin Rusch

Visual Language Models require substantial computational resources for inference due to the additional input tokens needed to represent visual information. However, these visual tokens often contain redundant and unimportant information,…

Computer Vision and Pattern Recognition · Computer Science 2025-04-15 Mohamed Dhouib , Davide Buscaldi , Sonia Vanier , Aymen Shabou

The attention mechanism is becoming increasingly popular in Natural Language Processing (NLP) applications, showing superior performance than convolutional and recurrent architectures. However, attention becomes the compution bottleneck…

Hardware Architecture · Computer Science 2024-07-22 Hanrui Wang , Zhekai Zhang , Song Han

Leveraging long contexts is crucial for advanced AI systems, but attention computation poses a scalability challenge. While scaled dot-product attention (SDPA) exhibits token sparsity, i.e. only a few pivotal tokens significantly contribute…

Machine Learning · Computer Science 2025-06-05 Aditya Desai , Shuo Yang , Alejandro Cuadron , Matei Zaharia , Joseph E. Gonzalez , Ion Stoica

The quadratic time and memory complexity inherent to self-attention mechanisms, with respect to sequence length, presents a critical computational bottleneck in the training and deployment of large-scale Transformer-based language models.…

Machine Learning · Computer Science 2024-03-19 Praneeth Kacham , Vahab Mirrokni , Peilin Zhong

This paper investigates the limitations of the normalization in attention mechanisms. We begin with a theoretical framework that enables the identification of the model's selective ability and the geometric separation involved in token…

Machine Learning · Computer Science 2025-10-21 Timur Mudarisov , Mikhail Burtsev , Tatiana Petrova , Radu State

Large language models rely on attention mechanisms with a softmax activation. Yet the dominance of softmax over alternatives (e.g., component-wise or linear) remains poorly understood, and many theoretical works have focused on the…

Machine Learning · Computer Science 2026-02-27 O. Duranthon , P. Marion , C. Boyer , B. Loureiro , L. Zdeborová

Memory consumption of the Key-Value (KV) cache represents a major bottleneck for efficient large language model inference. While attention-score-based KV cache pruning shows promise, it faces critical practical limitations: attention scores…

Artificial Intelligence · Computer Science 2025-10-02 Alessio Devoto , Maximilian Jeblick , Simon Jégou

Current efficient fine-tuning methods (e.g., adapters, prefix-tuning, etc.) have optimized conditional text generation via training a small set of extra parameters of the neural language model, while freezing the rest for efficiency. While…

Computation and Language · Computer Science 2022-05-24 Marjan Ghazvininejad , Vladimir Karpukhin , Vera Gor , Asli Celikyilmaz

Despite their ubiquity in language generation, it remains unknown why truncation sampling heuristics like nucleus sampling are so effective. We provide a theoretical explanation for the effectiveness of the truncation sampling by proving…

Computation and Language · Computer Science 2023-10-04 Matthew Finlayson , John Hewitt , Alexander Koller , Swabha Swayamdipta , Ashish Sabharwal

Prompt-tuning is an emerging strategy to adapt large language models (LLM) to downstream tasks by learning a (soft-)prompt parameter from data. Despite its success in LLMs, there is limited theoretical understanding of the power of…

Machine Learning · Computer Science 2023-06-07 Samet Oymak , Ankit Singh Rawat , Mahdi Soltanolkotabi , Christos Thrampoulidis

Diffusion models have revolutionized generative tasks, especially in the domain of text-to-image synthesis; however, their iterative denoising process demands substantial computational resources. In this paper, we present a novel…

Computer Vision and Pattern Recognition · Computer Science 2025-02-04 Xinle Cheng , Zhuoming Chen , Zhihao Jia

KV cache pruning has emerged as a promising technique for reducing memory and computation costs in long-context auto-regressive generation. Existing methods for vision-language models (VLMs) typically rely on self-attention scores from…

Computer Vision and Pattern Recognition · Computer Science 2024-12-09 Xiaohuan Pei , Tao Huang , Chang Xu

Diffusion Transformers (DiT) have become the de-facto model for generating high-quality visual content like videos and images. A huge bottleneck is the attention mechanism where complexity scales quadratically with resolution and video…

Computer Vision and Pattern Recognition · Computer Science 2025-10-30 Ruichen Chen , Keith G. Mills , Liyao Jiang , Chao Gao , Di Niu

Attention mechanism is a central component of the transformer architecture which led to the phenomenal success of large language models. However, the theoretical principles underlying the attention mechanism are poorly understood,…

Machine Learning · Computer Science 2023-12-11 Davoud Ataee Tarzanagh , Yingcong Li , Xuechen Zhang , Samet Oymak

This paper introduces and evaluates two novel Hierarchical Attention Network models [Yang et al., 2016] - i) Hierarchical Pruned Attention Networks, which remove the irrelevant words and sentences from the classification process in order to…

Computation and Language · Computer Science 2020-04-10 João G. Ribeiro , Frederico S. Felisberto , Isabel C. Neto

Fine tuning has been regarded as a de facto approach for adapting large language models (LLMs) to downstream tasks, but the high training memory consumption inherited from LLMs makes this process inefficient. Among existing memory efficient…

Computation and Language · Computer Science 2026-01-28 Runjia Zeng , Qifan Wang , Qiang Guan , Ruixiang Tang , Lifu Huang , Zhenting Wang , Xueling Zhang , Cheng Han , Dongfang Liu

Softmax-based dot-product attention is a cornerstone of Transformer architectures, enabling remarkable capabilities such as in-context learning. However, as context lengths increase, a fundamental limitation of the softmax function emerges:…

Machine Learning · Computer Science 2026-02-12 Sai Surya Duvvuri , Nirmal Patel , Nilesh Gupta , Inderjit S. Dhillon

Deploying transformer models in practice is challenging due to their inference cost, which scales quadratically with input sequence length. To address this, we present a novel Learned Token Pruning (LTP) method which adaptively removes…

Computation and Language · Computer Science 2022-06-06 Sehoon Kim , Sheng Shen , David Thorsley , Amir Gholami , Woosuk Kwon , Joseph Hassoun , Kurt Keutzer

Large language models typically employ vocabularies of over 100k tokens, which creates a major computational bottleneck at the final linear projection layer when performing speculative decoding. Current methods for vocabulary pruning depend…

Computation and Language · Computer Science 2026-05-27 Zhiyang Chen , Daliang Xu , Yinyuan Zhang , Chenghua Wang , Mengwei Xu , Yun Ma