SEARCH · Searcharxiv

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 541 records · Page 30Linked to original sources

arXiv2024

A Contrastive Learning Scheme with Transformer Innate Patches

This paper presents Contrastive Transformer, a contrastive learning scheme using the Transformer innate patches. Contrastive Transformer enables existing contrastive learning techniques, often used for image classification, to benefit dense downstream prediction tasks such as semantic segmentation. The scheme performs supervised patch-level contrastive learning, selecting the patches based on the ground truth mask, subsequently used for hard-negative and hard-positive sampling. The scheme applies to all vision-transformer architectures, is easy to implement, and introduces minimal additional memory footprint. Additionally, the scheme removes the need for huge batch sizes, as each patch is treated as an image. We apply and test Contrastive Transformer for the case of aerial image segmentation, known for low-resolution data, large class imbalance, and similar semantic classes. We perform extensive experiments to show the efficacy of the Contrastive Transformer scheme on the ISPRS Potsdam aerial image segmentation dataset. Additionally, we show the generalizability of our scheme by applying it to multiple inherently different Transformer architectures. Ultimately, the results show a consistent increase in mean IoU across all classes.

cs.CV↗

arXiv2025

Permutation-Invariant Transformer Neural Architectures for Set-Based Indoor Localization Using Learned RSSI Embeddings

We propose a permutation-invariant neural architecture for indoor localization using RSSI scans from Wi-Fi access points. Each scan is modeled as an unordered set of (BSSID, RSSI) pairs, where BSSIDs are mapped to learned embeddings and concatenated with signal strength. These are processed by a Set Transformer, enabling the model to handle variable-length, sparse inputs while learning attention-based representations over access point relationships. We evaluate the model on a dataset collected across a campus environment consisting of six buildings. Results show that the model accurately recovers fine-grained spatial structure and maintains performance across physically distinct domains. In our experiments, a simple LSTM consistently outperformed all other models, achieving the lowest mean localization error across three tasks (E1 - E3), with average errors as low as 2.23 m. The Set Transformer performed competitively, ranking second in every experiment and outperforming the MLP, RNN, and basic attention models, particularly in scenarios involving multiple buildings (E2) and multiple floors (E3). Performance degraded most in E2, where signal conditions varied substantially across buildings, highlighting the importance of architectural robustness to domain diversity. This work demonstrates that set-based neural models are a natural fit for signal-based localization, offering a principled approach to handling sparse, unordered inputs in real-world positioning tasks.

cs.LG↗

arXiv2025

Model Hemorrhage and the Robustness Limits of Large Language Models

Large language models (LLMs) demonstrate strong performance across natural language processing tasks, yet undergo significant performance degradation when modified for deployment through quantization, pruning, or decoding strategy adjustments. We define this phenomenon as model hemorrhage - performance decline caused by parameter alterations and architectural changes. Through systematic analysis of various LLM frameworks, we identify key vulnerability patterns: layer expansion frequently disrupts attention mechanisms, compression techniques induce information loss cascades, and decoding adjustments amplify prediction divergences. Our investigation reveals transformer architectures exhibit inherent robustness thresholds that determine hemorrhage severity across modification types. We propose three mitigation strategies: gradient-aware pruning preserves critical weight pathways, dynamic quantization scaling maintains activation integrity, and decoding calibration aligns generation trajectories with original model distributions. This work establishes foundational metrics for evaluating model stability during adaptation, providing practical guidelines for maintaining performance while enabling efficient LLM deployment. Our findings advance understanding of neural network resilience under architectural transformations, particularly for large-scale language models.

cs.CL↗

arXiv2023

Equiformer: Equivariant Graph Attention Transformer for 3D Atomistic Graphs

Despite their widespread success in various domains, Transformer networks have yet to perform well across datasets in the domain of 3D atomistic graphs such as molecules even when 3D-related inductive biases like translational invariance and rotational equivariance are considered. In this paper, we demonstrate that Transformers can generalize well to 3D atomistic graphs and present Equiformer, a graph neural network leveraging the strength of Transformer architectures and incorporating SE(3)/E(3)-equivariant features based on irreducible representations (irreps). First, we propose a simple and effective architecture by only replacing original operations in Transformers with their equivariant counterparts and including tensor products. Using equivariant operations enables encoding equivariant information in channels of irreps features without complicating graph structures. With minimal modifications to Transformers, this architecture has already achieved strong empirical results. Second, we propose a novel attention mechanism called equivariant graph attention, which improves upon typical attention in Transformers through replacing dot product attention with multi-layer perceptron attention and including non-linear message passing. With these two innovations, Equiformer achieves competitive results to previous models on QM9, MD17 and OC20 datasets.

cs.LG↗

arXiv2022

Temporal Attention for Language Models

Pretrained language models based on the transformer architecture have shown great success in NLP. Textual training data often comes from the web and is thus tagged with time-specific information, but most language models ignore this information. They are trained on the textual data alone, limiting their ability to generalize temporally. In this work, we extend the key component of the transformer architecture, i.e., the self-attention mechanism, and propose temporal attention - a time-aware self-attention mechanism. Temporal attention can be applied to any transformer model and requires the input texts to be accompanied with their relevant time points. It allows the transformer to capture this temporal information and create time-specific contextualized word representations. We leverage these representations for the task of semantic change detection; we apply our proposed mechanism to BERT and experiment on three datasets in different languages (English, German, and Latin) that also vary in time, size, and genre. Our proposed model achieves state-of-the-art results on all the datasets.

cs.CL↗

arXiv2024

Gasformer: A Transformer-based Architecture for Segmenting Methane Emissions from Livestock in Optical Gas Imaging

Methane emissions from livestock, particularly cattle, significantly contribute to climate change. Effective methane emission mitigation strategies are crucial as the global population and demand for livestock products increase. We introduce Gasformer, a novel semantic segmentation architecture for detecting low-flow rate methane emissions from livestock, and controlled release experiments using optical gas imaging. We present two unique datasets captured with a FLIR GF77 OGI camera. Gasformer leverages a Mix Vision Transformer encoder and a Light-Ham decoder to generate multi-scale features and refine segmentation maps. Gasformer outperforms other state-of-the-art models on both datasets, demonstrating its effectiveness in detecting and segmenting methane plumes in controlled and real-world scenarios. On the livestock dataset, Gasformer achieves mIoU of 88.56%, surpassing other state-of-the-art models. Materials are available at: github.com/toqitahamid/Gasformer.

cs.CV↗

arXiv2021

Open-Domain Conversational Search Assistant with Transformers

Open-domain conversational search assistants aim at answering user questions about open topics in a conversational manner. In this paper we show how the Transformer architecture achieves state-of-the-art results in key IR tasks, leveraging the creation of conversational assistants that engage in open-domain conversational search with single, yet informative, answers. In particular, we propose an open-domain abstractive conversational search agent pipeline to address two major challenges: first, conversation context-aware search and second, abstractive search-answers generation. To address the first challenge, the conversation context is modeled with a query rewriting method that unfolds the context of the conversation up to a specific moment to search for the correct answers. These answers are then passed to a Transformer-based re-ranker to further improve retrieval performance. The second challenge, is tackled with recent Abstractive Transformer architectures to generate a digest of the top most relevant passages. Experiments show that Transformers deliver a solid performance across all tasks in conversational search, outperforming the best TREC CAsT 2019 baseline.

cs.IR↗

arXiv2023

Transformers are Universal Predictors

We find limits to the Transformer architecture for language modeling and show it has a universal prediction property in an information-theoretic sense. We further analyze performance in non-asymptotic data regimes to understand the role of various components of the Transformer architecture, especially in the context of data-efficient training. We validate our theoretical analysis with experiments on both synthetic and real datasets.

cs.LG↗

arXiv2025

Bypassing the Exponential Dependency: Looped Transformers Efficiently Learn In-context by Multi-step Gradient Descent

In-context learning has been recognized as a key factor in the success of Large Language Models (LLMs). It refers to the model's ability to learn patterns on the fly from provided in-context examples in the prompt during inference. Previous studies have demonstrated that the Transformer architecture used in LLMs can implement a single-step gradient descent update by processing in-context examples in a single forward pass. Recent work has further shown that, during in-context learning, a looped Transformer can implement multi-step gradient descent updates in forward passes. However, their theoretical results require an exponential number of in-context examples, $n = \exp(Ω(T))$, where $T$ is the number of loops or passes, to achieve a reasonably low error. In this paper, we study linear looped Transformers in-context learning on linear vector generation tasks. We show that linear looped Transformers can implement multi-step gradient descent efficiently for in-context learning. Our results demonstrate that as long as the input data has a constant condition number, e.g., $n = O(d)$, the linear looped Transformers can achieve a small error by multi-step gradient descent during in-context learning. Furthermore, our preliminary experiments validate our theoretical analysis. Our findings reveal that the Transformer architecture possesses a stronger in-context learning capability than previously understood, offering new insights into the mechanisms behind LLMs and potentially guiding the better design of efficient inference algorithms for LLMs.

cs.LG↗

arXiv2026

Dynamic Windowing in Transformers via Regime Incorporation for Financial Time Series

Financial time series exhibit non-stationary behavior, where the strength and extent of temporal dependencies vary across market regimes. Trending, low-volatility phases typically require long-range contextual information, whereas mean-reverting, high-volatility periods rely more heavily on short-term dynamics. Standard Transformer architectures, with fixed attention windows and static positional encodings, are therefore unable to adapt to such variations. In this work, we propose a regime-aware dynamic windowing framework that incorporates market regime information directly into the Transformer. We construct four generic regime signals from price series: volatility ratio, trend strength, local predictability ratio (LPR), and rolling autocorrelation. We incorporate these signals into the model through two mechanisms: (i) regime-augmented inputs to a standard Transformer architecture, and (ii) a modified attention layer that modulates attention weights using regime embeddings. Experiments on five S&P 500 stocks show consistent improvements across five evaluation metrics, demonstrating that regime-aware dynamic windowing enhances both interpretability and predictive performance in financial forecasting tasks.

cs.CE↗

arXiv2024

Sparse-VQ Transformer: An FFN-Free Framework with Vector Quantization for Enhanced Time Series Forecasting

Time series analysis is vital for numerous applications, and transformers have become increasingly prominent in this domain. Leading methods customize the transformer architecture from NLP and CV, utilizing a patching technique to convert continuous signals into segments. Yet, time series data are uniquely challenging due to significant distribution shifts and intrinsic noise levels. To address these two challenges,we introduce the Sparse Vector Quantized FFN-Free Transformer (Sparse-VQ). Our methodology capitalizes on a sparse vector quantization technique coupled with Reverse Instance Normalization (RevIN) to reduce noise impact and capture sufficient statistics for forecasting, serving as an alternative to the Feed-Forward layer (FFN) in the transformer architecture. Our FFN-free approach trims the parameter count, enhancing computational efficiency and reducing overfitting. Through evaluations across ten benchmark datasets, including the newly introduced CAISO dataset, Sparse-VQ surpasses leading models with a 7.84% and 4.17% decrease in MAE for univariate and multivariate time series forecasting, respectively. Moreover, it can be seamlessly integrated with existing transformer-based models to elevate their performance.

cs.LG↗

arXiv2026

Multi-source Transfer Learning of Time Series with a Shapelet-based Distance Measure

Transfer learning is an effective technique for addressing data scarcity in deep learning for time series classification, but its success depends on the selection of source datasets. Conventional transferability estimation methods are often computationally expensive, as they require fully pre-training a model on each potential source dataset to assess its suitability. This paper introduces a novel, training-free source selection method named Shapelet Matching. Our approach first identifies discriminative shapelets from the target and potential source datasets. Then, Shapelet Matching quantifies dataset similarity by comparing the extracted sets of shapelets. To mitigate the risk of negative transfer from selecting an unsuitable single source, we introduce a multi-source transfer learning method. We select several source datasets based on their shapelet-based similarity scores, combine them into a single multi-source dataset, and use this aggregated dataset for pre-training. The model is then fine-tuned on the target task. We evaluated our method on 128 datasets from the UCR Archive using both temporal CNN and Transformer architectures. The empirical results demonstrate that our multi-source pre-training reduces the risk of negative transfer on average. Shapelet Matching achieves the strongest performance for the CNN backbone and remains competitive for patch-based Transformer architectures, while avoiding the cost of pre-training a separate model for every candidate source.

cs.LG↗

arXiv2024

Towards Multi-modal Transformers in Federated Learning

Multi-modal transformers mark significant progress in different domains, but siloed high-quality data hinders their further improvement. To remedy this, federated learning (FL) has emerged as a promising privacy-preserving paradigm for training models without direct access to the raw data held by different clients. Despite its potential, a considerable research direction regarding the unpaired uni-modal clients and the transformer architecture in FL remains unexplored. To fill this gap, this paper explores a transfer multi-modal federated learning (MFL) scenario within the vision-language domain, where clients possess data of various modalities distributed across different datasets. We systematically evaluate the performance of existing methods when a transformer architecture is utilized and introduce a novel framework called Federated modality complementary and collaboration (FedCola) by addressing the in-modality and cross-modality gaps among clients. Through extensive experiments across various FL settings, FedCola demonstrates superior performance over previous approaches, offering new perspectives on future federated training of multi-modal transformers.

cs.CV↗

arXiv2026

A statistical perspective on transformers for small longitudinal cohort data

Modeling of longitudinal cohort data typically involves complex temporal dependencies between multiple variables. There, the transformer architecture, which has been highly successful in language and vision applications, allows us to account for the fact that the most recently observed time points in an individual's history may not always be the most important for the immediate future. This is achieved by assigning attention weights to observations of an individual based on a transformation of their values. One reason why these ideas have not yet been fully leveraged for longitudinal cohort data is that typically, large datasets are required. Therefore, we present a simplified transformer architecture that retains the core attention mechanism while reducing the number of parameters to be estimated, to be more suitable for small datasets with few time points. Guided by a statistical perspective on transformers, we use an autoregressive model as a starting point and incorporate attention as a kernel-based operation with temporal decay, where aggregation of multiple transformer heads, i.e. different candidate weighting schemes, is expressed as accumulating evidence on different types of underlying characteristics of individuals. This also enables a permutation-based statistical testing procedure for identifying contextual patterns. In a simulation study, the approach is shown to recover contextual dependencies even with a small number of individuals and time points. In an application to data from a resilience study, we identify temporal patterns in the dynamics of stress and mental health. This indicates that properly adapted transformers can not only achieve competitive predictive performance, but also uncover complex context dependencies in small data settings.

stat.ME↗

arXiv2023

BEL: A Bag Embedding Loss for Transformer enhances Multiple Instance Whole Slide Image Classification

Multiple Instance Learning (MIL) has become the predominant approach for classification tasks on gigapixel histopathology whole slide images (WSIs). Within the MIL framework, single WSIs (bags) are decomposed into patches (instances), with only WSI-level annotation available. Recent MIL approaches produce highly informative bag level representations by utilizing the transformer architecture's ability to model the dependencies between instances. However, when applied to high magnification datasets, problems emerge due to the large number of instances and the weak supervisory learning signal. To address this problem, we propose to additionally train transformers with a novel Bag Embedding Loss (BEL). BEL forces the model to learn a discriminative bag-level representation by minimizing the distance between bag embeddings of the same class and maximizing the distance between different classes. We evaluate BEL with the Transformer architecture TransMIL on two publicly available histopathology datasets, BRACS and CAMELYON17. We show that with BEL, TransMIL outperforms the baseline models on both datasets, thus contributing to the clinically highly relevant AI-based tumor classification of histological patient material.

cs.CV↗

arXiv2023

Transformers as Statisticians: Provable In-Context Learning with In-Context Algorithm Selection

Neural sequence models based on the transformer architecture have demonstrated remarkable \emph{in-context learning} (ICL) abilities, where they can perform new tasks when prompted with training and test examples, without any parameter update to the model. This work first provides a comprehensive statistical theory for transformers to perform ICL. Concretely, we show that transformers can implement a broad class of standard machine learning algorithms in context, such as least squares, ridge regression, Lasso, learning generalized linear models, and gradient descent on two-layer neural networks, with near-optimal predictive power on various in-context data distributions. Using an efficient implementation of in-context gradient descent as the underlying mechanism, our transformer constructions admit mild size bounds, and can be learned with polynomially many pretraining sequences. Building on these ``base'' ICL algorithms, intriguingly, we show that transformers can implement more complex ICL procedures involving \emph{in-context algorithm selection}, akin to what a statistician can do in real life -- A \emph{single} transformer can adaptively select different base ICL algorithms -- or even perform qualitatively different tasks -- on different input sequences, without any explicit prompting of the right algorithm or task. We both establish this in theory by explicit constructions, and also observe this phenomenon experimentally. In theory, we construct two general mechanisms for algorithm selection with concrete examples: pre-ICL testing, and post-ICL validation. As an example, we use the post-ICL validation mechanism to construct a transformer that can perform nearly Bayes-optimal ICL on a challenging task -- noisy linear models with mixed noise levels. Experimentally, we demonstrate the strong in-context algorithm selection capabilities of standard transformer architectures.

cs.LG↗

arXiv2025

Brain-inspired Action Generation with Spiking Transformer Diffusion Policy Model

Spiking Neural Networks (SNNs) has the ability to extract spatio-temporal features due to their spiking sequence. While previous research has primarily foucus on the classification of image and reinforcement learning. In our paper, we put forward novel diffusion policy model based on Spiking Transformer Neural Networks and Denoising Diffusion Probabilistic Model (DDPM): Spiking Transformer Modulate Diffusion Policy Model (STMDP), a new brain-inspired model for generating robot action trajectories. In order to improve the performance of this model, we develop a novel decoder module: Spiking Modulate De coder (SMD), which replaces the traditional Decoder module within the Transformer architecture. Additionally, we explored the substitution of DDPM with Denoising Diffusion Implicit Models (DDIM) in our frame work. We conducted experiments across four robotic manipulation tasks and performed ablation studies on the modulate block. Our model consistently outperforms existing Transformer-based diffusion policy method. Especially in Can task, we achieved an improvement of 8%. The proposed STMDP method integrates SNNs, dffusion model and Transformer architecture, which offers new perspectives and promising directions for exploration in brain-inspired robotics.

cs.RO↗

arXiv2025

VESTA: A Versatile SNN-Based Transformer Accelerator with Unified PEs for Multiple Computational Layers

Spiking Neural Networks (SNNs) and transformers represent two powerful paradigms in neural computation, known for their low power consumption and ability to capture feature dependencies, respectively. However, transformer architectures typically involve multiple types of computational layers, including linear layers for MLP modules and classification heads, convolution layers for tokenizers, and dot product computations for self-attention mechanisms. These diverse operations pose significant challenges for hardware accelerator design, and to our knowledge, there is not yet a hardware solution that leverages spike-form data from SNNs for transformer architectures. In this paper, we introduce VESTA, a novel hardware design that synergizes these technologies, presenting unified Processing Elements (PEs) capable of efficiently performing all three types of computations crucial to transformer structures. VESTA uniquely benefits from the spike-form outputs of the Spike Neuron Layers \cite{zhou2024spikformer}, simplifying multiplication operations by reducing them from handling two 8-bit integers to handling one 8-bit integer and a binary spike. This reduction enables the use of multiplexers in the PE module, significantly enhancing computational efficiency while maintaining the low-power advantage of SNNs. Experimental results show that the core area of VESTA is $0.844 mm^2$. It operates at 500MHz and is capable of real-time image classification at 30 fps.

cs.AR↗