English
Related papers

Related papers: Adaptive Computation Pruning for the Forgetting Tr…

200 papers

Large Vision-Language Models (LVLMs) suffer from prohibitive inference costs due to the massive number of visual tokens processed by the language decoder. Existing pruning methods often lead to significant performance degradation because…

Computer Vision and Pattern Recognition · Computer Science 2026-04-08 Jianwei Zhang , Chaoning Zhang , Sihan Cao , Wang Liu , Pengcheng Zheng , Jiaxin Huang , Caiyan Qin , Yalan Ye , Wei Dong , Yang Yang

Standard inference and training with transformer based architectures scale quadratically with input sequence length. This is prohibitively large for a variety of applications especially in web-page translation, query-answering etc.…

Computation and Language · Computer Science 2023-03-20 Lovish Madaan , Srinadh Bhojanapalli , Himanshu Jain , Prateek Jain

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

This paper presents a novel attention-based algorithm for achieving adaptive computation called DACT, which, unlike existing ones, is end-to-end differentiable. Our method can be used in conjunction with many networks; in particular, we…

Artificial Intelligence · Computer Science 2020-05-25 Cristobal Eyzaguirre , Alvaro Soto

Recently, foundation models based on Vision Transformers (ViTs) have become widely available. However, their fine-tuning process is highly resource-intensive, and it hinders their adoption in several edge or low-energy applications. To this…

Computer Vision and Pattern Recognition · Computer Science 2024-08-19 Alessio Devoto , Federico Alvetreti , Jary Pomponi , Paolo Di Lorenzo , Pasquale Minervini , Simone Scardapane

Recently, large language models (LLM) based on transformers are facing memory bottleneck issues due to KV cache, especially in long sequence handling. Previous researches proposed KV cache compression techniques that identify insignificant…

Computation and Language · Computer Science 2024-08-01 Hyun-rae Jo , Dongkun Shin

The Transformer architecture model, based on self-attention and multi-head attention, has achieved remarkable success in offline end-to-end Automatic Speech Recognition (ASR). However, self-attention and multi-head attention cannot be…

Computation and Language · Computer Science 2022-10-03 Chendong Zhao , Jianzong Wang , Wen qi Wei , Xiaoyang Qu , Haoqian Wang , Jing Xiao

In Federated Learning, a global model is learned by aggregating model updates computed at a set of independent client nodes, to reduce communication costs multiple gradient steps are performed at each node prior to aggregation. A key…

Machine Learning · Computer Science 2023-04-12 Gwen Legate , Lucas Caccia , Eugene Belilovsky

Class-incremental learning is a challenging problem, where the goal is to train a model that can classify data from an increasing number of classes over time. With the advancement of vision-language pre-trained models such as CLIP, they…

Computer Vision and Pattern Recognition · Computer Science 2024-07-22 Linlan Huang , Xusheng Cao , Haori Lu , Xialei Liu

The attention mechanism is a key computing kernel of Transformers, calculating pairwise correlations across the entire input sequence. The computing complexity and frequent memory access in computing self-attention put a huge burden on the…

Hardware Architecture · Computer Science 2024-10-31 Ashkan Moradifirouzabadi , Divya Sri Dodla , Mingu Kang

Transformer-based architectures traditionally employ softmax to compute attention weights, which produces dense distributions over all tokens in a sequence. While effective in many settings, this density has been shown to be detrimental for…

Computation and Language · Computer Science 2026-03-03 Pavlo Vasylenko , Hugo Pitorro , André F. T. Martins , Marcos Treviso

Deploying useful Long-Context Transformer Models (LCTMs) requires addressing two key challenges: (1) A growing memory footprint due to quadratic self-attention and linear KV-cache scaling in memory as sequence length increases; (2) the…

Computation and Language · Computer Science 2025-10-15 Baisub Lee , Sanghyun Byun , Mohanad Odema , Jung Guack , Jacob Song , Woo Seong Chung

CTC-based ASR systems face computational and memory bottlenecks in resource-limited environments. Traditional CTC decoders, requiring up to 90% of processing time in systems (e.g., wav2vec2-large on L4 GPUs), face inefficiencies due to…

Machine Learning · Computer Science 2025-10-13 Atul Shree , Harshith Jupuru

The self-attention mechanism is the key to the success of transformers in recent Large Language Models (LLMs). However, the quadratic computational cost $O(n^2)$ in the input sequence length $n$ is a notorious obstacle for further…

Machine Learning · Computer Science 2024-10-17 Yingyu Liang , Heshan Liu , Zhenmei Shi , Zhao Song , Zhuoyan Xu , Junze Yin

The ever-growing size of training datasets enhances the generalization capability of modern machine learning models but also incurs exorbitant computational costs. Existing data pruning approaches aim to accelerate training by removing…

Computer Vision and Pattern Recognition · Computer Science 2025-07-01 Dongyue Wu , Zilin Guo , Jialong Zuo , Nong Sang , Changxin Gao

Large Language Models utilizing reasoning techniques improve task performance but incur significant latency and token costs due to verbose generation. Existing automatic prompt optimization(APO) frameworks target task accuracy exclusively…

Computation and Language · Computer Science 2026-04-17 Deep Shah , Sanket Badhe , Nehal Kathrotia , Priyanka Tiwari

In this paper, we introduce a framework ARBEx, a novel attentive feature extraction framework driven by Vision Transformer with reliability balancing to cope against poor class distributions, bias, and uncertainty in the facial expression…

Computer Vision and Pattern Recognition · Computer Science 2024-12-10 Azmine Toushik Wasi , Karlo Šerbetar , Raima Islam , Taki Hasan Rafi , Dong-Kyu Chae

Optimization algorithms are essential for solving many real-world problems. However, challenges such as getting trapped in local minima and effectively balancing exploration and exploitation often limit their performance. This paper…

Artificial Intelligence · Computer Science 2025-09-23 Mahmood A. Jumaah , Yossra H. Ali , Tarik A. Rashid

Autoregressive Transformers are strong language models but incur O(T) complexity during per-token generation due to the self-attention mechanism. Recent work proposes kernel-based methods to approximate causal self-attention by replacing it…

Machine Learning · Computer Science 2022-10-11 Huanru Henry Mao

Transformers have become the cornerstone of modern large-scale language models, but their reliance on softmax attention poses a computational bottleneck at both training and inference. Recurrent models offer high efficiency, but compressing…

Computation and Language · Computer Science 2025-11-20 Xiuying Wei , Anunay Yadav , Razvan Pascanu , Caglar Gulcehre