English
Related papers

Related papers: Pit One Against Many: Leveraging Attention-head Em…

200 papers

Combining several embeddings typically improves performance in downstream tasks as different embeddings encode different information. It has been shown that even models using embeddings from transformers still benefit from the inclusion of…

Computation and Language · Computer Science 2021-11-01 Lukas Lange , Heike Adel , Jannik Strötgen , Dietrich Klakow

Pre-trained language models (PLM) have demonstrated their effectiveness for a broad range of information retrieval and natural language processing tasks. As the core part of PLM, multi-head self-attention is appealing for its ability to…

Computation and Language · Computer Science 2022-04-07 Shanshan Wang , Zhumin Chen , Zhaochun Ren , Huasheng Liang , Qiang Yan , Pengjie Ren

Mixture-of-Experts (MoE) networks have been proposed as an efficient way to scale up model capacity and implement conditional computing. However, the study of MoE components mostly focused on the feedforward layer in Transformer…

Computation and Language · Computer Science 2022-10-12 Xiaofeng Zhang , Yikang Shen , Zeyu Huang , Jie Zhou , Wenge Rong , Zhang Xiong

Answering multi-hop reasoning questions requires retrieving and synthesizing information from diverse sources. Language models (LMs) struggle to perform such reasoning consistently. We propose an approach to pinpoint and rectify multi-hop…

Computation and Language · Computer Science 2024-11-11 Mansi Sakarvadia

Recently, self-attention based models have achieved state-of-the-art performance in sequential recommendation task. Following the custom from language processing, most of these models rely on a simple positional embedding to exploit the…

Machine Learning · Computer Science 2020-08-24 Sung Min Cho , Eunhyeok Park , Sungjoo Yoo

Transformers have reshaped machine learning by utilizing attention mechanisms to capture complex patterns in large datasets, leading to significant improvements in performance. This success has contributed to the belief that "bigger means…

Machine Learning · Computer Science 2025-05-28 Hemanth Saratchandran , Damien Teney , Simon Lucey

Attention based Transformer architecture has enabled significant advances in the field of natural language processing. In addition to new pre-training techniques, recent improvements crucially rely on working with a relatively larger…

Machine Learning · Computer Science 2020-02-18 Srinadh Bhojanapalli , Chulhee Yun , Ankit Singh Rawat , Sashank J. Reddi , Sanjiv Kumar

While Transformer self-attention offers strong parallelism, the Key-Value (KV) cache grows linearly with sequence length and becomes a bottleneck for inference efficiency. Multi-head latent attention was recently developed to compress the…

Machine Learning · Computer Science 2025-11-04 Keqi Deng , Philip C. Woodland

In this work, we upgrade the multi-head attention mechanism, the core of the Transformer model, to improve efficiency while maintaining or surpassing the previous accuracy level. We show that multi-head attention can be expressed in the…

Computer Vision and Pattern Recognition · Computer Science 2025-12-02 Peng Jin , Bo Zhu , Li Yuan , Shuicheng Yan

Slim attention shrinks the context memory size by 2x for transformer models with MHA (multi-head attention), which can speed up inference by up to 2x for large context windows. Slim attention is an exact, mathematically identical…

Machine Learning · Computer Science 2025-06-04 Nils Graef , Andrew Wasielewski

Multi-Head Latent Attention (MLA), introduced in DeepSeek-V2, improves the efficiency of large language models by projecting query, key, and value tensors into a compact latent space. This architectural change reduces the KV-cache size and…

Hardware Architecture · Computer Science 2026-04-10 Robin Geens , Marian Verhelst

As large language models (LLMs) and visual language models (VLMs) grow in scale and application, attention mechanisms have become a central computational bottleneck due to their high memory and time complexity. While many efficient…

Machine Learning · Computer Science 2025-07-11 Zhengyu Tian , Anantha Padmanaban Krishna Kumar , Hemant Krishnakumar , Reza Rawassizadeh

Transformers have advanced the field of natural language processing (NLP) on a variety of important tasks. At the cornerstone of the Transformer architecture is the multi-head attention (MHA) mechanism which models pairwise interactions…

Computation and Language · Computer Science 2021-06-01 Lin Zheng , Zhiyong Wu , Lingpeng Kong

We propose novel attention architectures, Multi-matrix Factorization Attention (MFA) and MFA-Key-Reuse (MFA-KR). Existing variants for standard Multi-Head Attention (MHA), including SOTA methods like MLA, fail to maintain as strong…

Machine Learning · Computer Science 2025-01-15 Jingcheng Hu , Houyi Li , Yinmin Zhang , Zili Wang , Shuigeng Zhou , Xiangyu Zhang , Heung-Yeung Shum , Daxin Jiang

Two principles: the complementary principle and the consensus principle are widely acknowledged in the literature of multi-view learning. However, the current design of multi-head self-attention, an instance of multi-view learning,…

Computation and Language · Computer Science 2024-06-06 Tong Zheng , Bei Li , Huiwen Bao , Tong Xiao , Jingbo Zhu

The neural attention mechanism plays an important role in many natural language processing applications. In particular, the use of multi-head attention extends single-head attention by allowing a model to jointly attend information from…

Machine Learning · Computer Science 2020-11-03 Bang An , Jie Lyu , Zhenyi Wang , Chunyuan Li , Changwei Hu , Fei Tan , Ruiyi Zhang , Yifan Hu , Changyou Chen

Large language models (LLMs) have demonstrated exceptional performance across diverse natural language processing tasks. However, as the model size and the input sequence's length increase, the linearly increasing key-value (KV) cache…

Computation and Language · Computer Science 2025-07-29 Qingyun Jin , Xiaohui Song , Feng Zhou , Zengchang Qin

We explore Multi-Head FFN (MH-FFN) as a replacement of FFN in the Transformer architecture, motivated by the structural similarity between single-head attention and FFN. While multi-head mechanisms enhance expressivity in attention, naively…

Machine Learning · Computer Science 2025-12-09 Minshen Zhang , Xiang Hu , Jianguo Li , Wei Wu , Kewei Tu

Simultaneous machine translation models start generating a target sequence before they have encoded or read the source sequence. Recent approaches for this task either apply a fixed policy on a state-of-the art Transformer model, or a…

Computation and Language · Computer Science 2019-09-30 Xutai Ma , Juan Pino , James Cross , Liezl Puzon , Jiatao Gu

Multi-head latent attention (MLA) is designed to optimize KV cache memory through low-rank key-value joint compression. Rather than caching keys and values separately, MLA stores their compressed latent representations, reducing memory…

Computation and Language · Computer Science 2025-09-09 Guihong Li , Mehdi Rezagholizadeh , Mingyu Yang , Vikram Appia , Emad Barsoum