SEARCH · Search arXiv

Results for “transformer architecture”

Search indexed arXiv papers on artificial intelligence, large language models, computer vision and robotics. Read source abstracts and follow links to arXiv.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,297 records · Page 72Linked to original sources

arXiv2025

A Separable Self-attention Inspired by the State Space Model for Computer Vision

Mamba is an efficient State Space Model (SSM) with linear computational complexity. Although SSMs are not suitable for handling non-causal data, Vision Mamba (ViM) methods still demonstrate good performance in tasks such as image classification and object detection. Recent studies have shown that there is a rich theoretical connection between state space models and attention variants. We propose a novel separable self attention method, for the first time introducing some excellent design concepts of Mamba into separable self-attention. To ensure a fair comparison with ViMs, we introduce VMINet, a simple yet powerful prototype architecture, constructed solely by stacking our novel attention modules with the most basic down-sampling layers. Notably, VMINet differs significantly from the conventional Transformer architecture. Our experiments demonstrate that VMINet has achieved competitive results on image classification and high-resolution dense prediction tasks.Code is available at: https://github.com/yws-wxs/VMINet.

cs.CV↗

arXiv2025

FreCT: Frequency-augmented Convolutional Transformer for Robust Time Series Anomaly Detection

Time series anomaly detection is critical for system monitoring and risk identification, across various domains, such as finance and healthcare. However, for most reconstruction-based approaches, detecting anomalies remains a challenge due to the complexity of sequential patterns in time series data. On the one hand, reconstruction-based techniques are susceptible to computational deviation stemming from anomalies, which can lead to impure representations of normal sequence patterns. On the other hand, they often focus on the time-domain dependencies of time series, while ignoring the alignment of frequency information beyond the time domain. To address these challenges, we propose a novel Frequency-augmented Convolutional Transformer (FreCT). FreCT utilizes patch operations to generate contrastive views and employs an improved Transformer architecture integrated with a convolution module to capture long-term dependencies while preserving local topology information. The introduced frequency analysis based on Fourier transformation could enhance the model's ability to capture crucial characteristics beyond the time domain. To protect the training quality from anomalies and improve the robustness, FreCT deploys stop-gradient Kullback-Leibler (KL) divergence and absolute error to optimize consistency information in both time and frequency domains. Extensive experiments on four public datasets demonstrate that FreCT outperforms existing methods in identifying anomalies.

cs.LG↗

arXiv2025

Provable In-Context Learning of Nonlinear Regression with Transformers

The transformer architecture, which processes sequences of input tokens to produce outputs for query tokens, has revolutionized numerous areas of machine learning. A defining feature of transformers is their ability to perform previously unseen tasks using task specific prompts without updating parameters, a phenomenon known as in-context learning (ICL). Recent research has actively explored the training dynamics behind ICL, with much of the focus on relatively simple tasks such as linear regression and binary classification. To advance the theoretical understanding of ICL, this paper investigates more complex nonlinear regression tasks, aiming to uncover how transformers acquire in-context learning capabilities in these settings. We analyze the stage-wise dynamics of attention during training: attention scores between a query token and its target features grow rapidly in the early phase, then gradually converge to one, while attention to irrelevant features decays more slowly and exhibits oscillatory behavior. Our analysis introduces new proof techniques that explicitly characterize how the nature of general non-degenerate $L$-Lipschitz task functions affects attention weights. Specifically, we identify that the Lipschitz constant $L$ of nonlinear function classes as a key factor governing the convergence dynamics of transformers in ICL. Leveraging these insights, for two distinct regimes depending on whether $L$ is below or above a threshold, we derive different time bounds to guarantee near-zero prediction error. Notably, despite the convergence time depending on the underlying task functions, we prove that query tokens consistently attend to prompt tokens with highly relevant features at convergence, demonstrating the ICL capability of transformers for unseen functions.

cs.LG↗

arXiv2025

Quantum Neural Network Architectures for Multivariate Time-Series Forecasting

In this paper, we address the challenge of multivariate time-series forecasting using quantum machine learning techniques. We introduce adaptation strategies that extend variational quantum circuit models, traditionally limited to univariate data, toward the multivariate setting, exploring both purely quantum and hybrid quantum-classical formulations. First, we extend and benchmark several VQC-based and hybrid architectures to systematically evaluate their capacity to model cross-variable dependencies. Second, building upon these foundations, we introduce the iQTransformer, a novel quantum transformer architecture that integrates a quantum self-attention mechanism within the iTransformer framework, enabling a quantum-native representation of inter-variable relationships. Third, we provide a comprehensive empirical evaluation on both synthetic and real-world datasets, showing that quantum-based models may achieve competitive or superior forecasting accuracy with fewer trainable parameters and faster convergence than state-of-the-art classical and quantum baselines in some cases. These contributions highlight the potential of quantum-enhanced architectures as efficient and scalable tools for advancing multivariate time-series forecasting.

quant-ph↗

arXiv2023

Enhancing Phenotype Recognition in Clinical Notes Using Large Language Models: PhenoBCBERT and PhenoGPT

We hypothesize that large language models (LLMs) based on the transformer architecture can enable automated detection of clinical phenotype terms, including terms not documented in the HPO. In this study, we developed two types of models: PhenoBCBERT, a BERT-based model, utilizing Bio+Clinical BERT as its pre-trained model, and PhenoGPT, a GPT-based model that can be initialized from diverse GPT models, including open-source versions such as GPT-J, Falcon, and LLaMA, as well as closed-source versions such as GPT-3 and GPT-3.5. We compared our methods with PhenoTagger, a recently developed HPO recognition tool that combines rule-based and deep learning methods. We found that our methods can extract more phenotype concepts, including novel ones not characterized by HPO. We also performed case studies on biomedical literature to illustrate how new phenotype information can be recognized and extracted. We compared current BERT-based versus GPT-based models for phenotype tagging, in multiple aspects including model architecture, memory usage, speed, accuracy, and privacy protection. We also discussed the addition of a negation step and an HPO normalization layer to the transformer models for improved HPO term tagging. In conclusion, PhenoBCBERT and PhenoGPT enable the automated discovery of phenotype terms from clinical notes and biomedical literature, facilitating automated downstream tasks to derive new biological insights on human diseases.

q-bio.QM↗

arXiv2021

A novel time-frequency Transformer based on self-attention mechanism and its application in fault diagnosis of rolling bearings

The scope of data-driven fault diagnosis models is greatly extended through deep learning (DL). However, the classical convolution and recurrent structure have their defects in computational efficiency and feature representation, while the latest Transformer architecture based on attention mechanism has not yet been applied in this field. To solve these problems, we propose a novel time-frequency Transformer (TFT) model inspired by the massive success of vanilla Transformer in sequence processing. Specially, we design a fresh tokenizer and encoder module to extract effective abstractions from the time-frequency representation (TFR) of vibration signals. On this basis, a new end-to-end fault diagnosis framework based on time-frequency Transformer is presented in this paper. Through the case studies on bearing experimental datasets, we construct the optimal Transformer structure and verify its fault diagnosis performance. The superiority of the proposed method is demonstrated in comparison with the benchmark models and other state-of-the-art methods.

cs.AI↗

arXiv2020

Self-attention encoding and pooling for speaker recognition

The computing power of mobile devices limits the end-user applications in terms of storage size, processing, memory and energy consumption. These limitations motivate researchers for the design of more efficient deep models. On the other hand, self-attention networks based on Transformer architecture have attracted remarkable interests due to their high parallelization capabilities and strong performance on a variety of Natural Language Processing (NLP) applications. Inspired by the Transformer, we propose a tandem Self-Attention Encoding and Pooling (SAEP) mechanism to obtain a discriminative speaker embedding given non-fixed length speech utterances. SAEP is a stack of identical blocks solely relied on self-attention and position-wise feed-forward networks to create vector representation of speakers. This approach encodes short-term speaker spectral features into speaker embeddings to be used in text-independent speaker verification. We have evaluated this approach on both VoxCeleb1 & 2 datasets. The proposed architecture is able to outperform the baseline x-vector, and shows competitive performance to some other benchmarks based on convolutions, with a significant reduction in model size. It employs 94%, 95%, and 73% less parameters compared to ResNet-34, ResNet-50, and x-vector, respectively. This indicates that the proposed fully attention based architecture is more efficient in extracting time-invariant features from speaker utterances.

eess.AS↗

arXiv2025

Graph-Aware Isomorphic Attention for Adaptive Dynamics in Transformers

We present an approach to modifying Transformer architectures by integrating graph-aware relational reasoning into the attention mechanism, merging concepts from graph neural networks and language modeling. Building on the inherent connection between attention and graph theory, we reformulate the Transformer's attention mechanism as a graph operation and propose Graph-Aware Isomorphic Attention. This method leverages advanced graph modeling strategies, including Graph Isomorphism Networks (GIN) and Principal Neighborhood Aggregation (PNA), to enrich the representation of relational structures. Our approach captures complex dependencies and generalizes across tasks, as evidenced by a reduced generalization gap and improved learning performance. Additionally, we expand the concept of graph-aware attention to introduce Sparse GIN-Attention, a fine-tuning approach that employs sparse GINs. By interpreting attention matrices as sparse adjacency graphs, this technique enhances the adaptability of pre-trained foundational models with minimal computational overhead, endowing them with graph-aware capabilities. Sparse GIN-Attention fine-tuning achieves improved training dynamics and better generalization compared to alternative methods like low-rank adaption (LoRA). We discuss latent graph-like structures within traditional attention mechanisms, offering a new lens through which Transformers can be understood. By evolving Transformers as hierarchical GIN models for relational reasoning. This perspective suggests profound implications for foundational model development, enabling the design of architectures that dynamically adapt to both local and global dependencies. Applications in bioinformatics, materials science, language modeling, and beyond could benefit from this synthesis of relational and sequential data modeling, setting the stage for interpretable and generalizable modeling strategies.

cs.LG↗

arXiv2025

Circuit Complexity Bounds for Visual Autoregressive Model

Understanding the expressive ability of a specific model is essential for grasping its capacity limitations. Recently, several studies have established circuit complexity bounds for Transformer architecture. Besides, the Visual AutoRegressive (VAR) model has risen to be a prominent method in the field of image generation, outperforming previous techniques, such as Diffusion Transformers, in generating high-quality images. We investigate the circuit complexity of the VAR model and establish a bound in this study. Our primary result demonstrates that the VAR model is equivalent to a simulation by a uniform $\mathsf{TC}^0$ threshold circuit with hidden dimension $d \leq O(n)$ and $\mathrm{poly}(n)$ precision. This is the first study to rigorously highlight the limitations in the expressive power of VAR models despite their impressive performance. We believe our findings will offer valuable insights into the inherent constraints of these models and guide the development of more efficient and expressive architectures in the future.

stat.ML↗

arXiv2025

ViT-LCA: A Neuromorphic Approach for Vision Transformers

The recent success of Vision Transformers has generated significant interest in attention mechanisms and transformer architectures. Although existing methods have proposed spiking self-attention mechanisms compatible with spiking neural networks, they often face challenges in effective deployment on current neuromorphic platforms. This paper introduces a novel model that combines vision transformers with the Locally Competitive Algorithm (LCA) to facilitate efficient neuromorphic deployment. Our experiments show that ViT-LCA achieves higher accuracy on ImageNet-1K dataset while consuming significantly less energy than other spiking vision transformer counterparts. Furthermore, ViT-LCA's neuromorphic-friendly design allows for more direct mapping onto current neuromorphic architectures.

cs.NE↗

arXiv2025

Scaling Linear Attention with Sparse State Expansion

The Transformer architecture, despite its widespread success, struggles with long-context scenarios due to quadratic computation and linear memory growth. While various linear attention variants mitigate these efficiency constraints by compressing context into fixed-size states, they often degrade performance in tasks such as in-context retrieval and reasoning. To address this limitation and achieve more effective context compression, we propose two key innovations. First, we introduce a row-sparse update formulation for linear attention by conceptualizing state updating as information classification. This enables sparse state updates via softmax-based top-$k$ hard classification, thereby extending receptive fields and reducing inter-class interference. Second, we present Sparse State Expansion (SSE) within the sparse framework, which expands the contextual state into multiple partitions, effectively decoupling parameter size from state capacity while maintaining the sparse classification paradigm. Supported by efficient parallelized implementations, our design achieves effective classification and highly discriminative state representations. We extensively validate SSE in both pure linear and hybrid (SSE-H) architectures across language modeling, in-context retrieval, and mathematical reasoning benchmarks. SSE demonstrates strong retrieval performance and scales favorably with state size. Moreover, after reinforcement learning (RL) training, our 2B SSE-H model achieves state-of-the-art mathematical reasoning performance among small reasoning models, scoring 64.5 on AIME24 and 50.2 on AIME25, significantly outperforming similarly sized open-source Transformers. These results highlight SSE as a promising and efficient architecture for long-context modeling.

cs.LG↗

arXiv2022

A Bird's-Eye Tutorial of Graph Attention Architectures

Graph Neural Networks (GNNs) have shown tremendous strides in performance for graph-structured problems especially in the domains of natural language processing, computer vision and recommender systems. Inspired by the success of the transformer architecture, there has been an ever-growing body of work on attention variants of GNNs attempting to advance the state of the art in many of these problems. Incorporating "attention" into graph mining has been viewed as a way to overcome the noisiness, heterogenity and complexity associated with graph-structured data as well as to encode soft-inductive bias. It is hence crucial and advantageous to study these variants from a bird's-eye view to assess their strengths and weaknesses. We provide a systematic and focused tutorial centered around attention based GNNs in a hope to benefit researchers dealing with graph-structured problems. Our tutorial looks at GNN variants from the point of view of the attention function and iteratively builds the reader's understanding of different graph attention variants.

cs.LG↗

arXiv2026

FANTF: Fuzzy attention network based transformers for time series analysis

This paper introduces Fuzzy Attention Network-Based Transformer (FANTF), an approach for time series forecasting, classification, and anomaly detection. FANTF combines fuzzy logic concepts with transformer architectures through a novel fuzzy attention mechanism. The proposed mechanism is inspired by Type-2 fuzzy set theory. It uses a learnable Gaussian uncertainty perturbation to model uncertainty in attention relevance scores. This makes it possible for the model to accurately capture uncertainty in temporal relationships better than normal deterministic attention. Transformers capacity to record complex temporal patterns and multivariate connections is enhanced by FANTF. The approach is both architecture-agnostic and task-agnostic. The proposed fuzzy attention module can be used as a direct replacement for the standard attention layer. To evaluate its effectiveness, FANTF is integrated into five transformer backbones: Transformer, Informer, PatchTST, Crossformer, and iTransformer. The fuzzy attention mechanism produces softer and more diverse attention distributions. It provides uncertainty-aware interpretations through attention heatmap visualizations. Experimental evaluations on some real-life datasets reveal that FANTF enhances the forecasting performance, classification, and anomaly detection tasks over traditional transformer-based models. FANTF achieves statistically significant and competitive performance improvements over standard transformer-based baselines on most benchmark datasets (p < 0.05, Wilcoxon signed-rank test), observed under noisy and non-stationary time series conditions.

cs.LG↗

arXiv2026

Registers Matter for Pixel-Space Diffusion Transformers

Vision Transformers (ViTs) are known to exhibit high-norm patch-token outliers that degrade feature map quality, a problem effectively mitigated by register tokens. As diffusion models increasingly adopt transformer architectures and move toward pixel-space training, they become closer in form to ViTs, raising the question of whether register tokens are also useful for Diffusion Transformers (DiTs). In this work, we show that DiTs differ from ViTs in a key respect: they do not exhibit patch-token outliers but still benefit from registers. Interestingly, registers are more effective in pixel-space DiTs than in latent-space DiTs. By analyzing intermediate representations, we find that register tokens produce cleaner feature maps at high noise levels, which may contribute to their effectiveness in pixel-space generation. We further observe that recent pixel-space DiT architectures implicitly incorporate register-like mechanisms, which may partially account for their strong empirical performance. Motivated by these observations, we propose Register Guidance, a technique that amplifies the contribution of register tokens responsible for improving visual structure and coherence.

cs.CV↗

arXiv2021

Knowledge-Aware Language Model Pretraining

How much knowledge do pretrained language models hold? Recent research observed that pretrained transformers are adept at modeling semantics but it is unclear to what degree they grasp human knowledge, or how to ensure they do so. In this paper we incorporate knowledge-awareness in language model pretraining without changing the transformer architecture, inserting explicit knowledge layers, or adding external storage of semantic information. Rather, we simply signal the existence of entities to the input of the transformer in pretraining, with an entity-extended tokenizer; and at the output, with an additional entity prediction task. Our experiments show that solely by adding these entity signals in pretraining, significantly more knowledge is packed into the transformer parameters: we observe improved language modeling accuracy, factual correctness in LAMA knowledge probing tasks, and semantics in the hidden representations through edge probing.We also show that our knowledge-aware language model (KALM) can serve as a drop-in replacement for GPT-2 models, significantly improving downstream tasks like zero-shot question-answering with no task-related training.

cs.CL↗

arXiv2020

Sub-Linear Memory: How to Make Performers SLiM

The Transformer architecture has revolutionized deep learning on sequential data, becoming ubiquitous in state-of-the-art solutions for a wide variety of applications. Yet vanilla Transformers are notoriously resource-expensive, requiring $O(L^2)$ in serial time and memory as functions of input length $L$. Recent works proposed various linear self-attention mechanisms, scaling only as $O(L)$ for serial computation. We perform a thorough analysis of recent Transformer mechanisms with linear self-attention, Performers, in terms of overall computational complexity. We observe a remarkable computational flexibility: forward and backward propagation can be performed with no approximations using sublinear memory as a function of $L$ (in addition to negligible storage for the input sequence), at a cost of greater time complexity in the parallel setting. In the extreme case, a Performer consumes only $O(1)$ memory during training, and still requires $O(L)$ time. This discovered time-memory tradeoff can be used for training or, due to complete backward-compatibility, for fine-tuning on a low-memory device, e.g. a smartphone or an earlier-generation GPU, thus contributing towards decentralized and democratized deep learning.

cs.LG↗

arXiv2023

Revisiting Deep Learning Models for Tabular Data

The existing literature on deep learning for tabular data proposes a wide range of novel architectures and reports competitive results on various datasets. However, the proposed models are usually not properly compared to each other and existing works often use different benchmarks and experiment protocols. As a result, it is unclear for both researchers and practitioners what models perform best. Additionally, the field still lacks effective baselines, that is, the easy-to-use models that provide competitive performance across different problems. In this work, we perform an overview of the main families of DL architectures for tabular data and raise the bar of baselines in tabular DL by identifying two simple and powerful deep architectures. The first one is a ResNet-like architecture which turns out to be a strong baseline that is often missing in prior works. The second model is our simple adaptation of the Transformer architecture for tabular data, which outperforms other solutions on most tasks. Both models are compared to many existing architectures on a diverse set of tasks under the same training and tuning protocols. We also compare the best DL models with Gradient Boosted Decision Trees and conclude that there is still no universally superior solution.

cs.LG↗

arXiv2021

Stable, Fast and Accurate: Kernelized Attention with Relative Positional Encoding

The attention module, which is a crucial component in Transformer, cannot scale efficiently to long sequences due to its quadratic complexity. Many works focus on approximating the dot-then-exponentiate softmax function in the original attention, leading to sub-quadratic or even linear-complexity Transformer architectures. However, we show that these methods cannot be applied to more powerful attention modules that go beyond the dot-then-exponentiate style, e.g., Transformers with relative positional encoding (RPE). Since in many state-of-the-art models, relative positional encoding is used as default, designing efficient Transformers that can incorporate RPE is appealing. In this paper, we propose a novel way to accelerate attention calculation for Transformers with RPE on top of the kernelized attention. Based upon the observation that relative positional encoding forms a Toeplitz matrix, we mathematically show that kernelized attention with RPE can be calculated efficiently using Fast Fourier Transform (FFT). With FFT, our method achieves $\mathcal{O}(n\log n)$ time complexity. Interestingly, we further demonstrate that properly using relative positional encoding can mitigate the training instability problem of vanilla kernelized attention. On a wide range of tasks, we empirically show that our models can be trained from scratch without any optimization issues. The learned model performs better than many efficient Transformer variants and is faster than standard Transformer in the long-sequence regime.

cs.LG↗