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 1,279 records · Page 71Linked to original sources

arXiv2024

VST++: Efficient and Stronger Visual Saliency Transformer

While previous CNN-based models have exhibited promising results for salient object detection (SOD), their ability to explore global long-range dependencies is restricted. Our previous work, the Visual Saliency Transformer (VST), addressed this constraint from a transformer-based sequence-to-sequence perspective, to unify RGB and RGB-D SOD. In VST, we developed a multi-task transformer decoder that concurrently predicts saliency and boundary outcomes in a pure transformer architecture. Moreover, we introduced a novel token upsampling method called reverse T2T for predicting a high-resolution saliency map effortlessly within transformer-based structures. Building upon the VST model, we further propose an efficient and stronger VST version in this work, i.e. VST++. To mitigate the computational costs of the VST model, we propose a Select-Integrate Attention (SIA) module, partitioning foreground into fine-grained segments and aggregating background information into a single coarse-grained token. To incorporate 3D depth information with low cost, we design a novel depth position encoding method tailored for depth maps. Furthermore, we introduce a token-supervised prediction loss to provide straightforward guidance for the task-related tokens. We evaluate our VST++ model across various transformer-based backbones on RGB, RGB-D, and RGB-T SOD benchmark datasets. Experimental results show that our model outperforms existing methods while achieving a 25% reduction in computational costs without significant performance compromise. The demonstrated strong ability for generalization, enhanced performance, and heightened efficiency of our VST++ model highlight its potential.

cs.CV↗

arXiv2026

Hierarchical Shift Mixing -- Beyond Dense Attention in Transformers

Since the introduction of the Transformer architecture for large language models, the softmax-based attention layer has faced increasing scrutinity due to its quadratic-time computational complexity. Attempts have been made to replace it with less complex methods, at the cost of reduced performance in most cases. We introduce Hierarchical Shift Mixing (HSM), a general framework for token mixing that distributes pairwise token interactions across Transformer layers rather than computing them densely within each layer. HSM enables linear-time complexity while remaining agnostic to the specific mixing function. We show that even simple HSM variants achieve performance close to softmax attention, and that hybrid architectures combining HSM with softmax attention can outperform a GPT-style Transformer baseline while reducing computational cost during both training and inference.

cs.LG↗

arXiv2025

Sliding Window Attention Training for Efficient Large Language Models

Recent advances in transformer-based Large Language Models (LLMs) have demonstrated remarkable capabilities across various tasks. However, their quadratic computational complexity concerning sequence length remains a significant bottleneck for processing long documents. As a result, many efforts like sparse attention and state space models have been proposed to improve the efficiency of LLMs over long sequences. Though effective, these approaches compromise the performance or introduce structural complexity. This calls for a simple yet efficient model that preserves the fundamental Transformer architecture. To this end, we introduce SWAT, which enables efficient long-context handling via Sliding Window Attention Training. This paper first attributes the inefficiency of Transformers to the attention sink phenomenon resulting from the high variance of softmax operation. Then, we replace softmax with the sigmoid function and utilize a balanced ALiBi and Rotary Position Embedding for efficient information compression and retention. Experiments demonstrate that SWAT achieves SOTA performance compared with state-of-the-art linear recurrent architectures on eight benchmarks. Code is available at https://github.com/Fzkuji/swat-attention.

cs.CL↗

arXiv2023

Adaptive Spot-Guided Transformer for Consistent Local Feature Matching

Local feature matching aims at finding correspondences between a pair of images. Although current detector-free methods leverage Transformer architecture to obtain an impressive performance, few works consider maintaining local consistency. Meanwhile, most methods struggle with large scale variations. To deal with the above issues, we propose Adaptive Spot-Guided Transformer (ASTR) for local feature matching, which jointly models the local consistency and scale variations in a unified coarse-to-fine architecture. The proposed ASTR enjoys several merits. First, we design a spot-guided aggregation module to avoid interfering with irrelevant areas during feature aggregation. Second, we design an adaptive scaling module to adjust the size of grids according to the calculated depth information at fine stage. Extensive experimental results on five standard benchmarks demonstrate that our ASTR performs favorably against state-of-the-art methods. Our code will be released on https://astr2023.github.io.

cs.CV↗

arXiv2026

Ordinary Least Squares is a Special Case of Transformer

The statistical essence of the Transformer architecture has long remained elusive: Is it a universal approximator, or a neural network version of known computational algorithms? Through rigorous algebraic proof, we show that the latter better describes Transformer's basic nature: Ordinary Least Squares (OLS) is a special case of the single-layer Linear Transformer. Using the spectral decomposition of the empirical covariance matrix, we construct a specific parameter setting where the attention mechanism's forward pass becomes mathematically equivalent to the OLS closed-form projection. This means attention can solve the problem in one forward pass, not by iterating. Building upon this prototypical case, we further uncover a decoupled slow and fast memory mechanism within Transformers. Finally, the evolution from our established linear prototype to standard Transformers is discussed. This progression facilitates the transition of the Hopfield energy function from linear to exponential memory capacity, thereby establishing a clear continuity between modern deep architectures and classical statistical inference.

cs.LG↗

arXiv2026

FaceLiVTv2: An Improved Hybrid Architecture for Efficient Mobile Face Recognition

Lightweight face recognition is increasingly important for deployment on edge and mobile devices, where strict constraints on latency, memory, and energy consumption must be met alongside reliable accuracy. Although recent hybrid CNN-Transformer architectures have advanced global context modeling, striking an effective balance between recognition performance and computational efficiency remains an open challenge. In this work, we present FaceLiVTv2, an improved version of our FaceLiVT hybrid architecture designed for efficient global--local feature interaction in mobile face recognition. At its core is Lite MHLA, a lightweight global token interaction module that replaces the original multi-layer attention design with multi-head linear token projections and affine rescale transformations, reducing redundancy while preserving representational diversity across heads. We further integrate Lite MHLA into a unified RepMix block that coordinates local and global feature interactions and adopts global depthwise convolution for adaptive spatial aggregation in the embedding stage. Under our experimental setup, results on LFW, CA-LFW, CP-LFW, CFP-FP, AgeDB-30, and IJB show that FaceLiVTv2 consistently improves the accuracy-efficiency trade-off over existing lightweight methods. Notably, FaceLiVTv2 reduces mobile inference latency by 22% relative to FaceLiVTv1, achieves speedups of up to 30.8% over GhostFaceNets on mobile devices, and delivers 20-41% latency improvements over EdgeFace and KANFace across platforms while maintaining higher recognition accuracy. These results demonstrate that FaceLiVTv2 offers a practical and deployable solution for real-time face recognition. Code is available at https://github.com/novendrastywn/FaceLiVT.

cs.CV↗

arXiv2026

A Mathematical Explanation of Transformers

The Transformer architecture has revolutionized the field of sequence modeling and underpins the recent breakthroughs in large language models (LLMs). However, a comprehensive mathematical theory that explains its structure and operations remains elusive. In this work, we propose a novel continuous framework that rigorously interprets the Transformer as a discretization of a structured integro-differential equation. Within this formulation, the self-attention mechanism emerges naturally as a non-local integral operator, and layer normalization is characterized as a projection to a time-dependent constraint. This operator-theoretic and variational perspective offers a unified and interpretable foundation for understanding the architecture's core components, including attention, feedforward layers, and normalization. Our approach extends beyond previous theoretical analyses by embedding the entire Transformer operation in continuous domains for both token indices and feature dimensions. This leads to a principled and flexible framework that not only deepens on theoretical insight but also offers new directions for architecture design, analysis, and control-based interpretations. This new interpretation provides a step toward bridging the gap between deep learning architectures and continuous mathematical modeling, and contributes a foundational perspective to the ongoing development of interpretable and theoretically grounded neural network models.

cs.LG↗

arXiv2026

Time-Delayed Transformers for Data-Driven Modeling of Low-Dimensional Dynamics

We propose the time-delayed transformer (TD-TF), a simplified transformer architecture for data-driven modeling of unsteady spatio-temporal dynamics. TD-TF bridges linear operator-based methods and deep sequence models by showing that a single-layer, single-head transformer can be interpreted as a nonlinear generalization of time-delayed dynamic mode decomposition (TD-DMD). The architecture is deliberately minimal, consisting of one self-attention layer with a single query per prediction and one feedforward layer, resulting in linear computational complexity in sequence length and a small parameter count. Numerical experiments demonstrate that TD-TF matches the performance of strong linear baselines on near-linear systems, while significantly outperforming them in nonlinear and chaotic regimes, where it accurately captures long-term dynamics. Validation studies on synthetic signals, unsteady aerodynamics, the Lorenz '63 system, and a reaction-diffusion model show that TD-TF preserves the interpretability and efficiency of linear models while providing substantially enhanced expressive power for complex dynamics.

cs.LG↗

arXiv2019

A Call for Prudent Choice of Subword Merge Operations in Neural Machine Translation

Most neural machine translation systems are built upon subword units extracted by methods such as Byte-Pair Encoding (BPE) or wordpiece. However, the choice of number of merge operations is generally made by following existing recipes. In this paper, we conduct a systematic exploration on different numbers of BPE merge operations to understand how it interacts with the model architecture, the strategy to build vocabularies and the language pair. Our exploration could provide guidance for selecting proper BPE configurations in the future. Most prominently: we show that for LSTM-based architectures, it is necessary to experiment with a wide range of different BPE operations as there is no typical optimal BPE configuration, whereas for Transformer architectures, smaller BPE size tends to be a typically optimal choice. We urge the community to make prudent choices with subword merge operations, as our experiments indicate that a sub-optimal BPE configuration alone could easily reduce the system performance by 3-4 BLEU points.

cs.CL↗

arXiv2020

Enhancing the Locality and Breaking the Memory Bottleneck of Transformer on Time Series Forecasting

Time series forecasting is an important problem across many domains, including predictions of solar plant energy output, electricity consumption, and traffic jam situation. In this paper, we propose to tackle such forecasting problem with Transformer [1]. Although impressed by its performance in our preliminary study, we found its two major weaknesses: (1) locality-agnostics: the point-wise dot-product self-attention in canonical Transformer architecture is insensitive to local context, which can make the model prone to anomalies in time series; (2) memory bottleneck: space complexity of canonical Transformer grows quadratically with sequence length $L$, making directly modeling long time series infeasible. In order to solve these two issues, we first propose convolutional self-attention by producing queries and keys with causal convolution so that local context can be better incorporated into attention mechanism. Then, we propose LogSparse Transformer with only $O(L(\log L)^{2})$ memory cost, improving forecasting accuracy for time series with fine granularity and strong long-term dependencies under constrained memory budget. Our experiments on both synthetic data and real-world datasets show that it compares favorably to the state-of-the-art.

cs.LG↗

arXiv2021

Boosting Objective Scores of a Speech Enhancement Model by MetricGAN Post-processing

The Transformer architecture has demonstrated a superior ability compared to recurrent neural networks in many different natural language processing applications. Therefore, our study applies a modified Transformer in a speech enhancement task. Specifically, positional encoding in the Transformer may not be necessary for speech enhancement, and hence, it is replaced by convolutional layers. To further improve the perceptual evaluation of the speech quality (PESQ) scores of enhanced speech, the L_1 pre-trained Transformer is fine-tuned using a MetricGAN framework. The proposed MetricGAN can be treated as a general post-processing module to further boost the objective scores of interest. The experiments were conducted using the data sets provided by the organizer of the Deep Noise Suppression (DNS) challenge. Experimental results demonstrated that the proposed system outperformed the challenge baseline, in both subjective and objective evaluations, with a large margin.

eess.AS↗

arXiv2021

Calliope -- A Polyphonic Music Transformer

The polyphonic nature of music makes the application of deep learning to music modelling a challenging task. On the other hand, the Transformer architecture seems to be a good fit for this kind of data. In this work, we present Calliope, a novel autoencoder model based on Transformers for the efficient modelling of multi-track sequences of polyphonic music. The experiments show that our model is able to improve the state of the art on musical sequence reconstruction and generation, with remarkably good results especially on long sequences.

cs.SD↗

arXiv2021

LCTR: On Awakening the Local Continuity of Transformer for Weakly Supervised Object Localization

Weakly supervised object localization (WSOL) aims to learn object localizer solely by using image-level labels. The convolution neural network (CNN) based techniques often result in highlighting the most discriminative part of objects while ignoring the entire object extent. Recently, the transformer architecture has been deployed to WSOL to capture the long-range feature dependencies with self-attention mechanism and multilayer perceptron structure. Nevertheless, transformers lack the locality inductive bias inherent to CNNs and therefore may deteriorate local feature details in WSOL. In this paper, we propose a novel framework built upon the transformer, termed LCTR (Local Continuity TRansformer), which targets at enhancing the local perception capability of global features among long-range feature dependencies. To this end, we propose a relational patch-attention module (RPAM), which considers cross-patch information on a global basis. We further design a cue digging module (CDM), which utilizes local features to guide the learning trend of the model for highlighting the weak local responses. Finally, comprehensive experiments are carried out on two widely used datasets, ie, CUB-200-2011 and ILSVRC, to verify the effectiveness of our method.

cs.CV↗

arXiv2022

Multiview Transformers for Video Recognition

Video understanding requires reasoning at multiple spatiotemporal resolutions -- from short fine-grained motions to events taking place over longer durations. Although transformer architectures have recently advanced the state-of-the-art, they have not explicitly modelled different spatiotemporal resolutions. To this end, we present Multiview Transformers for Video Recognition (MTV). Our model consists of separate encoders to represent different views of the input video with lateral connections to fuse information across views. We present thorough ablation studies of our model and show that MTV consistently performs better than single-view counterparts in terms of accuracy and computational cost across a range of model sizes. Furthermore, we achieve state-of-the-art results on six standard datasets, and improve even further with large-scale pretraining. Code and checkpoints are available at: https://github.com/google-research/scenic/tree/main/scenic/projects/mtv.

cs.CV↗

arXiv2023

Mitigating Bias in Visual Transformers via Targeted Alignment

As transformer architectures become increasingly prevalent in computer vision, it is critical to understand their fairness implications. We perform the first study of the fairness of transformers applied to computer vision and benchmark several bias mitigation approaches from prior work. We visualize the feature space of the transformer self-attention modules and discover that a significant portion of the bias is encoded in the query matrix. With this knowledge, we propose TADeT, a targeted alignment strategy for debiasing transformers that aims to discover and remove bias primarily from query matrix features. We measure performance using Balanced Accuracy and Standard Accuracy, and fairness using Equalized Odds and Balanced Accuracy Difference. TADeT consistently leads to improved fairness over prior work on multiple attribute prediction tasks on the CelebA dataset, without compromising performance.

cs.CV↗

arXiv2023

Large Sequence Models for Sequential Decision-Making: A Survey

Transformer architectures have facilitated the development of large-scale and general-purpose sequence models for prediction tasks in natural language processing and computer vision, e.g., GPT-3 and Swin Transformer. Although originally designed for prediction problems, it is natural to inquire about their suitability for sequential decision-making and reinforcement learning problems, which are typically beset by long-standing issues involving sample efficiency, credit assignment, and partial observability. In recent years, sequence models, especially the Transformer, have attracted increasing interest in the RL communities, spawning numerous approaches with notable effectiveness and generalizability. This survey presents a comprehensive overview of recent works aimed at solving sequential decision-making tasks with sequence models such as the Transformer, by discussing the connection between sequential decision-making and sequence modeling, and categorizing them based on the way they utilize the Transformer. Moreover, this paper puts forth various potential avenues for future research intending to improve the effectiveness of large sequence models for sequential decision-making, encompassing theoretical foundations, network architectures, algorithms, and efficient training systems. As this article has been accepted by the Frontiers of Computer Science, here is an early version, and the most up-to-date version can be found at https://journal.hep.com.cn/fcs/EN/10.1007/s11704-023-2689-5

cs.LG↗

arXiv2025

Multi-entity Video Transformers for Fine-Grained Video Representation Learning

The area of temporally fine-grained video representation learning focuses on generating frame-by-frame representations for temporally dense tasks, such as fine-grained action phase classification and frame retrieval. In this work, we advance the state-of-the-art for self-supervised models in this area by re-examining the design of transformer architectures for video representation learning. A key aspect of our approach is the improved sharing of scene information in the temporal pipeline by representing multiple salient entities per frame. Prior works use late-fusion architectures that reduce frames to a single-dimensional vector before modeling any cross-frame dynamics. In contrast, our Multi-entity Video Transformer (MV-Former) processes the frames as groups of entities represented as tokens linked across time. To achieve this, we propose a Learnable Spatial Token Pooling strategy to identify and extract features for multiple salient regions per frame. Through our experiments, we show that MV-Former outperforms previous self-supervised methods, and also surpasses some prior works that use additional supervision or training data. When combined with additional pre-training data from Kinetics-400, MV-Former achieves a further performance boost. Overall, our MV-Former achieves state-of-the-art results on multiple fine-grained video benchmarks and shows that parsing video scenes as collections of entities can enhance performance in video tasks.

cs.CV↗

arXiv2024

SalFoM: Dynamic Saliency Prediction with Video Foundation Models

Recent advancements in video saliency prediction (VSP) have shown promising performance compared to the human visual system, whose emulation is the primary goal of VSP. However, current state-of-the-art models employ spatio-temporal transformers trained on limited amounts of data, hindering generalizability adaptation to downstream tasks. The benefits of vision foundation models present a potential solution to improve the VSP process. However, adapting image foundation models to the video domain presents significant challenges in modeling scene dynamics and capturing temporal information. To address these challenges, and as the first initiative to design a VSP model based on video foundation models, we introduce SalFoM, a novel encoder-decoder video transformer architecture. Our model employs UnMasked Teacher (UMT) as feature extractor and presents a heterogeneous decoder which features a locality-aware spatio-temporal transformer and integrates local and global spatio-temporal information from various perspectives to produce the final saliency map. Our qualitative and quantitative experiments on the challenging VSP benchmark datasets of DHF1K, Hollywood-2 and UCF-Sports demonstrate the superiority of our proposed model in comparison with the state-of-the-art methods.

cs.CV↗