English
Related papers

Related papers: Adaptive Computation Depth via Learned Token Routi…

200 papers

Attention layers -- which map a sequence of inputs to a sequence of outputs -- are core building blocks of the Transformer architecture which has achieved significant breakthroughs in modern artificial intelligence. This paper presents a…

Machine Learning · Computer Science 2023-07-24 Hengyu Fu , Tianyu Guo , Yu Bai , Song Mei

The rigid, uniform allocation of computation in standard Transformer (TF) architectures can limit their efficiency and scalability, particularly for large-scale models and long sequences. Addressing this, we introduce Subjective Depth…

Machine Learning · Computer Science 2025-11-27 Frederico Wieser , Martin Benfeghoul , Haitham Bou Ammar , Jun Wang , Zafeirios Fountas

The Transformer architecture, underpinned by the Multi-Head Attention (MHA) mechanism, has become the de facto standard for state-of-the-art models in artificial intelligence. However, the quadratic computational complexity of MHA with…

Machine Learning · Computer Science 2025-10-03 Adam Filipek

DeepSeek Sparse Attention (DSA) sets the state of the art for fine-grained inference-time sparse attention by introducing a learned token-wise indexer that scores every prefix token and selects the most relevant ones for the main attention.…

Machine Learning · Computer Science 2026-05-11 Ruijie Zhou , Fanxu Meng , Yufei Xu , Tongxuan Liu , Guangming Lu , Muhan Zhang , Wenjie Pei

We introduce Monte-Carlo Attention (MCA), a randomized approximation method for reducing the computational cost of self-attention mechanisms in Transformer architectures. MCA exploits the fact that the importance of each token in an input…

Machine Learning · Computer Science 2022-02-01 Hyunjun Kim , JeongGil Ko

There is an increasing interest in the application of deep learning architectures to tabular data. One of the state-of-the-art solutions is TabTransformer which incorporates an attention mechanism to better track relationships between…

Machine Learning · Computer Science 2022-01-04 Radostin Cholakov , Todor Kolev

Conditional depth execution routes a subset of tokens through a lightweight cheap FFN while the remainder execute the standard full FFN at each controlled layer. The central difficulty is gate training: the gate decision must propagate…

Machine Learning · Computer Science 2026-04-21 Qingwei Lin

Despite progress across a broad range of applications, Transformers have limited success in systematic generalization. The situation is especially frustrating in the case of algorithmic tasks, where they often fail to find intuitive…

Machine Learning · Computer Science 2022-05-06 Róbert Csordás , Kazuki Irie , Jürgen Schmidhuber

While Transformer architecture excel at modeling long-range dependencies contributing to its widespread adoption in vision tasks the quadratic complexity of softmax-based attention mechanisms imposes a major bottleneck, particularly when…

Computer Vision and Pattern Recognition · Computer Science 2026-03-10 Yuan Cao , Dong Wang

A widely cited result by Dong et al. (2021) showed that Transformers built from self-attention alone, without skip connections or feed-forward layers, suffer from rapid rank collapse: all token representations converge to a single…

Machine Learning · Computer Science 2026-04-28 Giansalvo Cirrincione

DDCL-Attention is a prototype-based readout layer for transformer encoders that replaces simple pooling methods, such as mean pooling or class tokens, with a learned compression mechanism. It uses a small set of global prototype vectors and…

Machine Learning · Computer Science 2026-04-07 Giansalvo Cirrincione , Rahul Ranjeev Kumar

Multimodal Large Language Models (MLLMs) struggle with continual learning, often suffering from catastrophic forgetting when adapting to sequential tasks. We introduce a routing-based architecture that integrates new capabilities while…

Machine Learning · Computer Science 2026-04-08 Jay Mohta , Kenan Emir Ak , Gwang Lee , Dimitrios Dimitriadis , Yan Xu , Mingwei Shen

Multi-Head Latent Attention (MLA), introduced in DeepSeek-V2, compresses key-value states into a low-rank latent vector, caching only this vector to reduce memory. In tensor parallelism (TP), however, attention heads are computed across…

Machine Learning · Computer Science 2025-08-26 Xiaojuan Tang , Fanxu Meng , Pingzhi Tang , Yuxuan Wang , Di Yin , Xing Sun , Muhan Zhang

The transformer is a fundamental building block in deep learning, and the attention mechanism is the transformer's core component. Self-supervised speech representation learning (SSRL) represents a popular use-case for the transformer…

Sound · Computer Science 2024-03-19 Jianbo Ma , Siqi Pan , Deepak Chandran , Andrea Fanelli , Richard Cartwright

Sequence modelling requires determining which past tokens are causally relevant from the context and their importance: a process inherent to the attention layers in transformers, yet whose underlying learned mechanisms remain poorly…

Machine Learning · Computer Science 2026-04-14 Francesco D'Angelo , Nicolas Flammarion

With the rapid growth in model size, fine-tuning the large pre-trained language model has become increasingly difficult due to its extensive memory usage. Previous works usually focus on reducing the number of trainable parameters in the…

Transformers, renowned for their self-attention mechanism, have achieved state-of-the-art performance across various tasks in natural language processing, computer vision, time-series modeling, etc. However, one of the challenges with deep…

Machine Learning · Computer Science 2024-11-04 Jeongwhan Choi , Hyowon Wi , Jayoung Kim , Yehjin Shin , Kookjin Lee , Nathaniel Trask , Noseong Park

Transformers have revolutionized natural language processing, but their quadratic complexity with respect to sequence length remains a fundamental bottleneck for long-range modeling. While sparse attention mechanisms like RingAttention…

Computation and Language · Computer Science 2026-03-31 Dong Liu , Yanxuan Yu

Intermediate training of pre-trained transformer-based language models on domain-specific data leads to substantial gains for downstream tasks. To increase efficiency and prevent catastrophic forgetting alleviated from full domain-adaptive…

Computation and Language · Computer Science 2023-05-23 Chia-Chien Hung , Lukas Lange , Jannik Strötgen

Transformer-based deep learning models have achieved state-of-the-art performance across numerous language and vision tasks. While the self-attention mechanism, a core component of transformers, has proven capable of handling complex data…

Machine Learning · Computer Science 2025-08-05 Laziz Abdullaev , Tan M. Nguyen
‹ Prev 1 8 9 10 Next ›