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 1,207 records · Page 67Linked to original sources

arXiv2025

Contextually Guided Transformers via Low-Rank Adaptation

Large Language Models (LLMs) based on Transformers excel at text processing, but their reliance on prompts for specialized behavior introduces computational overhead. We propose a modification to a Transformer architecture that eliminates the need for explicit prompts by learning to encode context into the model's weights. Our Contextually Guided Transformer (CGT) model maintains a contextual summary at each sequence position, allowing it to update the weights on the fly based on the preceding context. This approach enables the model to self-specialize, effectively creating a tailored model for processing information following a given prefix. We demonstrate the effectiveness of our method on synthetic in-context learning tasks and language modeling benchmarks. Furthermore, we introduce techniques for enhancing the interpretability of the learned contextual representations, drawing connections to Variational Autoencoders and promoting smoother, more consistent context encoding. This work offers a novel direction for efficient and adaptable language modeling by integrating context directly into the model's architecture.

cs.LG↗

arXiv2022

TransBTSV2: Towards Better and More Efficient Volumetric Segmentation of Medical Images

Transformer, benefiting from global (long-range) information modeling using self-attention mechanism, has been successful in natural language processing and computer vision recently. Convolutional Neural Networks, capable of capturing local features, are difficult to model explicit long-distance dependencies from global feature space. However, both local and global features are crucial for dense prediction tasks, especially for 3D medical image segmentation. In this paper, we present the further attempt to exploit Transformer in 3D CNN for 3D medical image volumetric segmentation and propose a novel network named TransBTSV2 based on the encoder-decoder structure. Different from TransBTS, the proposed TransBTSV2 is not limited to brain tumor segmentation (BTS) but focuses on general medical image segmentation, providing a stronger and more efficient 3D baseline for volumetric segmentation of medical images. As a hybrid CNN-Transformer architecture, TransBTSV2 can achieve accurate segmentation of medical images without any pre-training, possessing the strong inductive bias as CNNs and powerful global context modeling ability as Transformer. With the proposed insight to redesign the internal structure of Transformer block and the introduced Deformable Bottleneck Module to capture shape-aware local details, a highly efficient architecture is achieved with superior performance. Extensive experimental results on four medical image datasets (BraTS 2019, BraTS 2020, LiTS 2017 and KiTS 2019) demonstrate that TransBTSV2 achieves comparable or better results compared to the state-of-the-art methods for the segmentation of brain tumor, liver tumor as well as kidney tumor. Code will be publicly available at https://github.com/Wenxuan-1119/TransBTS.

eess.IV↗

arXiv2022

Efficient Attention-free Video Shift Transformers

This paper tackles the problem of efficient video recognition. In this area, video transformers have recently dominated the efficiency (top-1 accuracy vs FLOPs) spectrum. At the same time, there have been some attempts in the image domain which challenge the necessity of the self-attention operation within the transformer architecture, advocating the use of simpler approaches for token mixing. However, there are no results yet for the case of video recognition, where the self-attention operator has a significantly higher impact (compared to the case of images) on efficiency. To address this gap, in this paper, we make the following contributions: (a) we construct a highly efficient \& accurate attention-free block based on the shift operator, coined Affine-Shift block, specifically designed to approximate as closely as possible the operations in the MHSA block of a Transformer layer. Based on our Affine-Shift block, we construct our Affine-Shift Transformer and show that it already outperforms all existing shift/MLP--based architectures for ImageNet classification. (b) We extend our formulation in the video domain to construct Video Affine-Shift Transformer (VAST), the very first purely attention-free shift-based video transformer. (c) We show that VAST significantly outperforms recent state-of-the-art transformers on the most popular action recognition benchmarks for the case of models with low computational and memory footprint. Code will be made available.

cs.CV↗

arXiv2026

Component-Aware Self-Speculative Decoding in Hybrid Language Models

Speculative decoding accelerates autoregressive inference by drafting candidate tokens with a fast model and verifying them in parallel with the target. Self-speculative methods avoid the need for an external drafter but have been studied exclusively in homogeneous Transformer architectures. We introduce component-aware self-speculative decoding, the first method to exploit the internal architectural heterogeneity of hybrid language models, isolating the SSM/linear-attention subgraph as a zero-cost internal draft. We evaluate this on two architecturally distinct hybrid families: Falcon-H1 (parallel: Mamba-2 + attention per layer) and Qwen3.5 (sequential: interleaved linear and attention layers), with a pure Transformer control (Qwen2.5). Parallel hybrids achieve acceptance rates of alpha = 0.68 at draft length k=2 under greedy decoding, while sequential hybrids yield only alpha = 0.038 -- an 18x gap attributable to how each architecture integrates its components. The property is scale-invariant: Falcon-H1 at 3B reproduces the rates observed at 0.5B. We further show that perplexity degradation from a companion ablation study predicts speculative viability without running speculative decoding: a 3.15x ratio (Falcon) maps to alpha = 0.37 at k=4, while 81.96x (Qwen) maps to alpha = 0.019. For sequential hybrids, generic LayerSkip achieves 12x higher acceptance rates than the component-aware strategy. The composition pattern of hybrid models -- not merely the presence of alternative components -- determines whether component-level self-speculation is viable.

cs.CL↗

arXiv2026

A Reproducibility Study of Partial Residual Ablations in Pre-LN Transformers

Residual connections are a fundamental component of transformer architectures, yet the roles of the attention and feed-forward residual pathways remain poorly understood when considered independently. This paper presents a reproducibility study of partial residual ablations in Pre-LN GPT-style transformers trained at two scales (10M and 124M parameters). I compare four architectural configurations by selectively removing the attention residual connection, the feed-forward residual connection, or both. At 10M scale, a controlled 8-seed deterministic sweep shows a clear asymmetry: removing the attention residual (FFNOnly) reaches the No-Residual collapse regime (3.350 +/- 0.002), whereas removing the feed-forward residual (AttnOnly) remains well separated from it (1.580 +/- 0.003). A subsequent controlled 124M sweep establishes that the direction of this asymmetry persists under matched seeds: across five seeds each, AttnOnly reaches 6.037 +/- 0.355 versus 7.569 +/- 0.001 for FFNOnly, with no overlap between the observed ranges. AttnOnly nevertheless remains substantially degraded relative to FullResidual (4.577 +/- 0.012) and exhibits markedly greater seed sensitivity at 124M. During the investigation, I identified and corrected an experimental measurement confound in runtime gain scaling and retained an intermediate reproduction failure rather than excluding it. I propose cross-position routing through self-attention as a falsifiable hypothesis for the observed asymmetry; no clean causal test of this mechanism has yet been completed. To support reproducibility, I release the source code, experiment configurations, training logs, and experimental results, including intermediate non-reproducing runs.

cs.LG↗

arXiv2023

Fast Training of NMT Model with Data Sorting

The Transformer model has revolutionized Natural Language Processing tasks such as Neural Machine Translation, and many efforts have been made to study the Transformer architecture, which increased its efficiency and accuracy. One potential area for improvement is to address the computation of empty tokens that the Transformer computes only to discard them later, leading to an unnecessary computational burden. To tackle this, we propose an algorithm that sorts translation sentence pairs based on their length before batching, minimizing the waste of computing power. Since the amount of sorting could violate the independent and identically distributed (i.i.d) data assumption, we sort the data partially. In experiments, we apply the proposed method to English-Korean and English-Luganda language pairs for machine translation and show that there are gains in computational time while maintaining the performance. Our method is independent of architectures, so that it can be easily integrated into any training process with flexible data lengths.

cs.CL↗

arXiv2022

Local Slot Attention for Vision-and-Language Navigation

Vision-and-language navigation (VLN), a frontier study aiming to pave the way for general-purpose robots, has been a hot topic in the computer vision and natural language processing community. The VLN task requires an agent to navigate to a goal location following natural language instructions in unfamiliar environments. Recently, transformer-based models have gained significant improvements on the VLN task. Since the attention mechanism in the transformer architecture can better integrate inter- and intra-modal information of vision and language. However, there exist two problems in current transformer-based models. 1) The models process each view independently without taking the integrity of the objects into account. 2) During the self-attention operation in the visual modality, the views that are spatially distant can be inter-weaved with each other without explicit restriction. This kind of mixing may introduce extra noise instead of useful information. To address these issues, we propose 1) A slot-attention based module to incorporate information from segmentation of the same object. 2) A local attention mask mechanism to limit the visual attention span. The proposed modules can be easily plugged into any VLN architecture and we use the Recurrent VLN-Bert as our base model. Experiments on the R2R dataset show that our model has achieved the state-of-the-art results.

cs.CV↗

arXiv2020

Multi-task Learning with Multi-head Attention for Multi-choice Reading Comprehension

Multiple-choice Machine Reading Comprehension (MRC) is an important and challenging Natural Language Understanding (NLU) task, in which a machine must choose the answer to a question from a set of choices, with the question placed in context of text passages or dialog. In the last a couple of years the NLU field has been revolutionized with the advent of models based on the Transformer architecture, which are pretrained on massive amounts of unsupervised data and then fine-tuned for various supervised learning NLU tasks. Transformer models have come to dominate a wide variety of leader-boards in the NLU field; in the area of MRC, the current state-of-the-art model on the DREAM dataset (see[Sunet al., 2019]) fine tunes Albert, a large pretrained Transformer-based model, and addition-ally combines it with an extra layer of multi-head attention between context and question-answer[Zhuet al., 2020].The purpose of this note is to document a new state-of-the-art result in the DREAM task, which is accomplished by, additionally, performing multi-task learning on two MRC multi-choice reading comprehension tasks (RACE and DREAM).

cs.CL↗

arXiv2021

The Piano Inpainting Application

Autoregressive models are now capable of generating high-quality minute-long expressive MIDI piano performances. Even though this progress suggests new tools to assist music composition, we observe that generative algorithms are still not widely used by artists due to the limited control they offer, prohibitive inference times or the lack of integration within musicians' workflows. In this work, we present the Piano Inpainting Application (PIA), a generative model focused on inpainting piano performances, as we believe that this elementary operation (restoring missing parts of a piano performance) encourages human-machine interaction and opens up new ways to approach music composition. Our approach relies on an encoder-decoder Linear Transformer architecture trained on a novel representation for MIDI piano performances termed Structured MIDI Encoding. By uncovering an interesting synergy between Linear Transformers and our inpainting task, we are able to efficiently inpaint contiguous regions of a piano performance, which makes our model suitable for interactive and responsive A.I.-assisted composition. Finally, we introduce our freely-available Ableton Live PIA plugin, which allows musicians to smoothly generate or modify any MIDI clip using PIA within a widely-used professional Digital Audio Workstation.

cs.SD↗

arXiv2022

Towards Theoretically Inspired Neural Initialization Optimization

Automated machine learning has been widely explored to reduce human efforts in designing neural architectures and looking for proper hyperparameters. In the domain of neural initialization, however, similar automated techniques have rarely been studied. Most existing initialization methods are handcrafted and highly dependent on specific architectures. In this paper, we propose a differentiable quantity, named GradCosine, with theoretical insights to evaluate the initial state of a neural network. Specifically, GradCosine is the cosine similarity of sample-wise gradients with respect to the initialized parameters. By analyzing the sample-wise optimization landscape, we show that both the training and test performance of a network can be improved by maximizing GradCosine under gradient norm constraint. Based on this observation, we further propose the neural initialization optimization (NIO) algorithm. Generalized from the sample-wise analysis into the real batch setting, NIO is able to automatically look for a better initialization with negligible cost compared with the training time. With NIO, we improve the classification performance of a variety of neural architectures on CIFAR-10, CIFAR-100, and ImageNet. Moreover, we find that our method can even help to train large vision Transformer architecture without warmup.

cs.LG↗

arXiv2023

Performance Analysis of Transformer Based Models (BERT, ALBERT and RoBERTa) in Fake News Detection

Fake news is fake material in a news media format but is not processed properly by news agencies. The fake material can provoke or defame significant entities or individuals or potentially even for the personal interests of the creators, causing problems for society. Distinguishing fake news and real news is challenging due to limited of domain knowledge and time constraints. According to the survey, the top three areas most exposed to hoaxes and misinformation by residents are in Banten, DKI Jakarta and West Java. The model of transformers is referring to an approach in the field of artificial intelligence (AI) in natural language processing utilizing the deep learning architectures. Transformers exercise a powerful attention mechanism to process text in parallel and produce rich and contextual word representations. A previous study indicates a superior performance of a transformer model known as BERT over and above non transformer approach. However, some studies suggest the performance can be improved with the use of improved BERT models known as ALBERT and RoBERTa. However, the modified BERT models are not well explored for detecting fake news in Bahasa Indonesia. In this research, we explore those transformer models and found that ALBERT outperformed other models with 87.6% accuracy, 86.9% precision, 86.9% F1-score, and 174.5 run-time (s/epoch) respectively. Source code available at: https://github.com/Shafna81/fakenewsdetection.git

cs.CL↗

arXiv2023

MCUFormer: Deploying Vision Transformers on Microcontrollers with Limited Memory

Due to the high price and heavy energy consumption of GPUs, deploying deep models on IoT devices such as microcontrollers makes significant contributions for ecological AI. Conventional methods successfully enable convolutional neural network inference of high resolution images on microcontrollers, while the framework for vision transformers that achieve the state-of-the-art performance in many vision applications still remains unexplored. In this paper, we propose a hardware-algorithm co-optimizations method called MCUFormer to deploy vision transformers on microcontrollers with extremely limited memory, where we jointly design transformer architecture and construct the inference operator library to fit the memory resource constraint. More specifically, we generalize the one-shot network architecture search (NAS) to discover the optimal architecture with highest task performance given the memory budget from the microcontrollers, where we enlarge the existing search space of vision transformers by considering the low-rank decomposition dimensions and patch resolution for memory reduction. For the construction of the inference operator library of vision transformers, we schedule the memory buffer during inference through operator integration, patch embedding decomposition, and token overwriting, allowing the memory buffer to be fully utilized to adapt to the forward pass of the vision transformer. Experimental results demonstrate that our MCUFormer achieves 73.62\% top-1 accuracy on ImageNet for image classification with 320KB memory on STM32F746 microcontroller. Code is available at https://github.com/liangyn22/MCUFormer.

cs.CV↗

arXiv2025

AlgoFormer: An Efficient Transformer Framework with Algorithmic Structures

Besides natural language processing, transformers exhibit extraordinary performance in solving broader applications, including scientific computing and computer vision. Previous works try to explain this from the expressive power and capability perspectives that standard transformers are capable of performing some algorithms. To empower transformers with algorithmic capabilities and motivated by the recently proposed looped transformer, we design a novel transformer framework, dubbed Algorithm Transformer (abbreviated as AlgoFormer). We provide an insight that efficient transformer architectures can be designed by leveraging prior knowledge of tasks and the underlying structure of potential algorithms. Compared with the standard transformer and vanilla looped transformer, the proposed AlgoFormer can perform efficiently in algorithm representation in some specific tasks. In particular, inspired by the structure of human-designed learning algorithms, our transformer framework consists of a pre-transformer that is responsible for task preprocessing, a looped transformer for iterative optimization algorithms, and a post-transformer for producing the desired results after post-processing. We provide theoretical evidence of the expressive power of the AlgoFormer in solving some challenging problems, mirroring human-designed algorithms. Furthermore, some theoretical and empirical results are presented to show that the designed transformer has the potential to perform algorithm representation and learning. Experimental results demonstrate the empirical superiority of the proposed transformer in that it outperforms the standard transformer and vanilla looped transformer in some specific tasks. An extensive experiment on real language tasks (e.g., neural machine translation of German and English, and text classification) further validates the expressiveness and effectiveness of AlgoFormer.

cs.LG↗

arXiv2024

UPS: Efficiently Building Foundation Models for PDE Solving via Cross-Modal Adaptation

We present Unified PDE Solvers (UPS), a data- and compute-efficient approach to developing unified neural operators for diverse families of spatiotemporal PDEs from various domains, dimensions, and resolutions. UPS embeds different PDEs into a shared representation space and processes them using a FNO-transformer architecture. Rather than training the network from scratch, which is data-demanding and computationally expensive, we warm-start the transformer from pretrained LLMs and perform explicit alignment to reduce the modality gap while improving data and compute efficiency. The cross-modal UPS achieves state-of-the-art results on a wide range of 1D and 2D PDE families from PDEBench, outperforming existing unified models using 4 times less data and 26 times less compute. Meanwhile, it is capable of few-shot transfer to unseen PDE families and coefficients.

cs.LG↗

arXiv2024

Advancing Spiking Neural Networks towards Multiscale Spatiotemporal Interaction Learning

Recent advancements in neuroscience research have propelled the development of Spiking Neural Networks (SNNs), which not only have the potential to further advance neuroscience research but also serve as an energy-efficient alternative to Artificial Neural Networks (ANNs) due to their spike-driven characteristics. However, previous studies often neglected the multiscale information and its spatiotemporal correlation between event data, leading SNN models to approximate each frame of input events as static images. We hypothesize that this oversimplification significantly contributes to the performance gap between SNNs and traditional ANNs. To address this issue, we have designed a Spiking Multiscale Attention (SMA) module that captures multiscale spatiotemporal interaction information. Furthermore, we developed a regularization method named Attention ZoneOut (AZO), which utilizes spatiotemporal attention weights to reduce the model's generalization error through pseudo-ensemble training. Our approach has achieved state-of-the-art results on mainstream neural morphology datasets. Additionally, we have reached a performance of 77.1% on the Imagenet-1K dataset using a 104-layer ResNet architecture enhanced with SMA and AZO. This achievement confirms the state-of-the-art performance of SNNs with non-transformer architectures and underscores the effectiveness of our method in bridging the performance gap between SNN models and traditional ANN models.

cs.CV↗

arXiv2024

Pathformer: Recursive Path Query Encoding for Complex Logical Query Answering

Complex Logical Query Answering (CLQA) over incomplete knowledge graphs is a challenging task. Recently, Query Embedding (QE) methods are proposed to solve CLQA by performing multi-hop logical reasoning. However, most of them only consider historical query context information while ignoring future information, which leads to their failure to capture the complex dependencies behind the elements of a query. In recent years, the transformer architecture has shown a strong ability to model long-range dependencies between words. The bidirectional attention mechanism proposed by the transformer can solve the limitation of these QE methods regarding query context. Still, as a sequence model, it is difficult for the transformer to model complex logical queries with branch structure computation graphs directly. To this end, we propose a neural one-point embedding method called Pathformer based on the tree-like computation graph, i.e., query computation tree. Specifically, Pathformer decomposes the query computation tree into path query sequences by branches and then uses the transformer encoder to recursively encode these path query sequences to obtain the final query embedding. This allows Pathformer to fully utilize future context information to explicitly model the complex interactions between various parts of the path query. Experimental results show that Pathformer outperforms existing competitive neural QE methods, and we found that Pathformer has the potential to be applied to non-one-point embedding space.

cs.LG↗

arXiv2026

Large-Scale Multi-omic Biosequence Transformers for Modeling Protein-Nucleic Acid Interactions

The transformer architecture has revolutionized bioinformatics and driven progress in the understanding and prediction of the properties of biomolecules. To date, most biosequence transformers have been trained on single-omic data - either proteins or nucleic acids - and have seen incredible success in downstream tasks in each domain, with particularly noteworthy breakthroughs in protein structural modeling. However, single-omic pretraining limits the ability of these models to capture cross-modal interactions. Here we present OmniBioTE, the largest open-source multi-omic model trained on over 250 billion tokens of mixed protein and nucleic acid data. We show that despite only being trained on unlabeled sequence data, OmniBioTE learns joint representations mapping genes to their corresponding protein sequences. We further demonstrate that OmniBioTE achieves state-of-the-art results predicting the change in Gibbs free energy ({ΔG}) of the binding interaction between a given nucleic acid and protein. Remarkably, we show that multi-omic biosequence transformers emergently learn useful structural information without any a priori structural training, allowing us to predict which protein residues are most involved in the protein-nucleic acid binding interaction. Compared to single-omic controls trained with identical compute, OmniBioTE also demonstrates superior performance-per-FLOP across both multi-omic and single-omic benchmarks. Together, these results highlight the power of a unified modeling approach for biological sequences and establish OmniBioTE as a foundation model for multi-omic discovery.

cs.LG↗

arXiv2025

CMAWRNet: Multiple Adverse Weather Removal via a Unified Quaternion Neural Architecture

Images used in real-world applications such as image or video retrieval, outdoor surveillance, and autonomous driving suffer from poor weather conditions. When designing robust computer vision systems, removing adverse weather such as haze, rain, and snow is a significant problem. Recently, deep-learning methods offered a solution for a single type of degradation. Current state-of-the-art universal methods struggle with combinations of degradations, such as haze and rain-streak. Few algorithms have been developed that perform well when presented with images containing multiple adverse weather conditions. This work focuses on developing an efficient solution for multiple adverse weather removal using a unified quaternion neural architecture called CMAWRNet. It is based on a novel texture-structure decomposition block, a novel lightweight encoder-decoder quaternion transformer architecture, and an attentive fusion block with low-light correction. We also introduce a quaternion similarity loss function to preserve color information better. The quantitative and qualitative evaluation of the current state-of-the-art benchmarking datasets and real-world images shows the performance advantages of the proposed CMAWRNet compared to other state-of-the-art weather removal approaches dealing with multiple weather artifacts. Extensive computer simulations validate that CMAWRNet improves the performance of downstream applications such as object detection. This is the first time the decomposition approach has been applied to the universal weather removal task.

cs.CV↗