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 829 records · Page 46Linked to original sources

arXiv2024

MART: MultiscAle Relational Transformer Networks for Multi-agent Trajectory Prediction

Multi-agent trajectory prediction is crucial to autonomous driving and understanding the surrounding environment. Learning-based approaches for multi-agent trajectory prediction, such as primarily relying on graph neural networks, graph transformers, and hypergraph neural networks, have demonstrated outstanding performance on real-world datasets in recent years. However, the hypergraph transformer-based method for trajectory prediction is yet to be explored. Therefore, we present a MultiscAle Relational Transformer (MART) network for multi-agent trajectory prediction. MART is a hypergraph transformer architecture to consider individual and group behaviors in transformer machinery. The core module of MART is the encoder, which comprises a Pair-wise Relational Transformer (PRT) and a Hyper Relational Transformer (HRT). The encoder extends the capabilities of a relational transformer by introducing HRT, which integrates hyperedge features into the transformer mechanism, promoting attention weights to focus on group-wise relations. In addition, we propose an Adaptive Group Estimator (AGE) designed to infer complex group relations in real-world environments. Extensive experiments on three real-world datasets (NBA, SDD, and ETH-UCY) demonstrate that our method achieves state-of-the-art performance, enhancing ADE/FDE by 3.9%/11.8% on the NBA dataset. Code is available at https://github.com/gist-ailab/MART.

cs.LG↗

arXiv2024

M3: Mamba-assisted Multi-Circuit Optimization via MBRL with Effective Scheduling

Recent advancements in reinforcement learning (RL) for analog circuit optimization have demonstrated significant potential for improving sample efficiency and generalization across diverse circuit topologies and target specifications. However, there are challenges such as high computational overhead, the need for bespoke models for each circuit. To address them, we propose M3, a novel Model-based RL (MBRL) method employing the Mamba architecture and effective scheduling. The Mamba architecture, known as a strong alternative to the transformer architecture, enables multi-circuit optimization with distinct parameters and target specifications. The effective scheduling strategy enhances sample efficiency by adjusting crucial MBRL training parameters. To the best of our knowledge, M3 is the first method for multi-circuit optimization by leveraging both the Mamba architecture and a MBRL with effective scheduling. As a result, it significantly improves sample efficiency compared to existing RL methods.

cs.LG↗

arXiv2025

Lowering the Error Floor of Error Correction Code Transformer

With the success of transformer architectures across diverse applications, the error correction code transformer (ECCT) has gained significant attention for its superior decoding performance. In spite of its advantages, the error floor problem in ECCT decoding remains unexplored. We present the first investigation into this issue, revealing that ECCT encounters error floors, limiting its effectiveness in practical settings. To address this error floor problem, we adopt a hybrid decoding framework that integrates ECCT with conventional hard decision decoders. Unlike prior hybrid decoding schemes, our key contribution lies in proposing a novel loss function that explicitly takes into account the interaction between ECCT and hard decision decoders during training. The proposed loss function guides ECCT to focus on residual errors that are not corrected by the hard decision stages, effectively lowering the error floor. Simulation results confirm that the hybrid decoder trained with the proposed loss function achieves substantial performance gains over standard ECCT in both the waterfall and the error floor regions.

cs.IT↗

arXiv2025

AbbIE: Autoregressive Block-Based Iterative Encoder for Efficient Sequence Modeling

We introduce the Autoregressive Block-Based Iterative Encoder (AbbIE), a novel recursive generalization of the encoder-only Transformer architecture, which achieves better perplexity than a standard Transformer and allows for the dynamic scaling of compute resources at test time. This simple, recursive approach is a complement to scaling large language model (LLM) performance through parameter and token counts. AbbIE performs its iterations in latent space, but unlike latent reasoning models, does not require a specialized dataset or training protocol. We show that AbbIE upward generalizes (ability to generalize to arbitrary iteration lengths) at test time by only using 2 iterations during train time, far outperforming alternative iterative methods. AbbIE's ability to scale its computational expenditure based on the complexity of the task gives it an up to \textbf{12\%} improvement in zero-shot in-context learning tasks versus other iterative and standard methods and up to 5\% improvement in language perplexity. The results from this study open a new avenue to Transformer performance scaling. We perform all of our evaluations on model sizes up to 350M parameters.

cs.LG↗

arXiv2025

Should We Ever Prefer Decision Transformer for Offline Reinforcement Learning?

In recent years, extensive work has explored the application of the Transformer architecture to reinforcement learning problems. Among these, Decision Transformer (DT) has gained particular attention in the context of offline reinforcement learning due to its ability to frame return-conditioned policy learning as a sequence modeling task. Most recently, Bhargava et al. (2024) provided a systematic comparison of DT with more conventional MLP-based offline RL algorithms, including Behavior Cloning (BC) and Conservative Q-Learning (CQL), and claimed that DT exhibits superior performance in sparse-reward and low-quality data settings. In this paper, through experimentation on robotic manipulation tasks (Robomimic) and locomotion benchmarks (D4RL), we show that MLP-based Filtered Behavior Cloning (FBC) achieves competitive or superior performance compared to DT in sparse-reward environments. FBC simply filters out low-performing trajectories from the dataset and then performs ordinary behavior cloning on the filtered dataset. FBC is not only very straightforward, but it also requires less training data and is computationally more efficient. The results therefore suggest that DT is not preferable for sparse-reward environments. From prior work, arguably, DT is also not preferable for dense-reward environments. Thus, we pose the question: Is DT ever preferable?

cs.AI↗

arXiv2025

GestureHYDRA: Semantic Co-speech Gesture Synthesis via Hybrid Modality Diffusion Transformer and Cascaded-Synchronized Retrieval-Augmented Generation

While increasing attention has been paid to co-speech gesture synthesis, most previous works neglect to investigate hand gestures with explicit and essential semantics. In this paper, we study co-speech gesture generation with an emphasis on specific hand gesture activation, which can deliver more instructional information than common body movements. To achieve this, we first build a high-quality dataset of 3D human body movements including a set of semantically explicit hand gestures that are commonly used by live streamers. Then we present a hybrid-modality gesture generation system GestureHYDRA built upon a hybrid-modality diffusion transformer architecture with novelly designed motion-style injective transformer layers, which enables advanced gesture modeling ability and versatile gesture operations. To guarantee these specific hand gestures can be activated, we introduce a cascaded retrieval-augmented generation strategy built upon a semantic gesture repository annotated for each subject and an adaptive audio-gesture synchronization mechanism, which substantially improves semantic gesture activation and production efficiency. Quantitative and qualitative experiments demonstrate that our proposed approach achieves superior performance over all the counterparts. The project page can be found at https://mumuwei.github.io/GestureHYDRA/.

cs.MM↗

arXiv2025

RS-CA-HSICT: A Residual and Spatial Channel Augmented CNN Transformer Framework for Monkeypox Detection

This work proposes a hybrid deep learning approach, namely Residual and Spatial Learning based Channel Augmented Integrated CNN-Transformer architecture, that leverages the strengths of CNN and Transformer towards enhanced MPox detection. The proposed RS-CA-HSICT framework is composed of an HSICT block, a residual CNN module, a spatial CNN block, and a CA, which enhances the diverse feature space, detailed lesion information, and long-range dependencies. The new HSICT module first integrates an abstract representation of the stem CNN and customized ICT blocks for efficient multihead attention and structured CNN layers with homogeneous (H) and structural (S) operations. The customized ICT blocks learn global contextual interactions and local texture extraction. Additionally, H and S layers learn spatial homogeneity and fine structural details by reducing noise and modeling complex morphological variations. Moreover, inverse residual learning enhances vanishing gradient, and stage-wise resolution reduction ensures scale invariance. Furthermore, the RS-CA-HSICT framework augments the learned HSICT channels with the TL-driven Residual and Spatial CNN maps for enhanced multiscale feature space capturing global and localized structural cues, subtle texture, and contrast variations. These channels, preceding augmentation, are refined through the Channel-Fusion-and-Attention block, which preserves discriminative channels while suppressing redundant ones, thereby enabling efficient computation. Finally, the spatial attention mechanism refines pixel selection to detect subtle patterns and intra-class contrast variations in Mpox. Experimental results on both the Kaggle benchmark and a diverse MPox dataset reported classification accuracy as high as 98.30% and an F1-score of 98.13%, which outperforms the existing CNNs and ViTs.

cs.CV↗

arXiv2024

Multimodal Metadata Assignment for Cultural Heritage Artifacts

We develop a multimodal classifier for the cultural heritage domain using a late fusion approach and introduce a novel dataset. The three modalities are Image, Text, and Tabular data. We based the image classifier on a ResNet convolutional neural network architecture and the text classifier on a multilingual transformer architecture (XML-Roberta). Both are trained as multitask classifiers and use the focal loss to handle class imbalance. Tabular data and late fusion are handled by Gradient Tree Boosting. We also show how we leveraged specific data models and taxonomy in a Knowledge Graph to create the dataset and to store classification results. All individual classifiers accurately predict missing properties in the digitized silk artifacts, with the multimodal approach providing the best results.

cs.CV↗

arXiv2024

Deciphering genomic codes using advanced NLP techniques: a scoping review

Objectives: The vast and complex nature of human genomic sequencing data presents challenges for effective analysis. This review aims to investigate the application of Natural Language Processing (NLP) techniques, particularly Large Language Models (LLMs) and transformer architectures, in deciphering genomic codes, focusing on tokenization, transformer models, and regulatory annotation prediction. The goal of this review is to assess data and model accessibility in the most recent literature, gaining a better understanding of the existing capabilities and constraints of these tools in processing genomic sequencing data. Methods: Following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, our scoping review was conducted across PubMed, Medline, Scopus, Web of Science, Embase, and ACM Digital Library. Studies were included if they focused on NLP methodologies applied to genomic sequencing data analysis, without restrictions on publication date or article type. Results: A total of 26 studies published between 2021 and April 2024 were selected for review. The review highlights that tokenization and transformer models enhance the processing and understanding of genomic data, with applications in predicting regulatory annotations like transcription-factor binding sites and chromatin accessibility. Discussion: The application of NLP and LLMs to genomic sequencing data interpretation is a promising field that can help streamline the processing of large-scale genomic data while also providing a better understanding of its complex structures. It has the potential to drive advancements in personalized medicine by offering more efficient and scalable solutions for genomic analysis. Further research is also needed to discuss and overcome current limitations, enhancing model transparency and applicability.

q-bio.GN↗

arXiv2023

DeFTAN-II: Efficient Multichannel Speech Enhancement with Subgroup Processing

In this work, we present DeFTAN-II, an efficient multichannel speech enhancement model based on transformer architecture and subgroup processing. Despite the success of transformers in speech enhancement, they face challenges in capturing local relations, reducing the high computational complexity, and lowering memory usage. To address these limitations, we introduce subgroup processing in our model, combining subgroups of locally emphasized features with other subgroups containing original features. The subgroup processing is implemented in several blocks of the proposed network. In the proposed split dense blocks extracting spatial features, a pair of subgroups is sequentially concatenated and processed by convolution layers to effectively reduce the computational complexity and memory usage. For the F- and T-transformers extracting temporal and spectral relations, we introduce cross-attention between subgroups to identify relationships between locally emphasized and non-emphasized features. The dual-path feedforward network then aggregates attended features in terms of the gating of local features processed by dilated convolutions. Through extensive comparisons with state-of-the-art multichannel speech enhancement models, we demonstrate that DeFTAN-II with subgroup processing outperforms existing methods at significantly lower computational complexity. Moreover, we evaluate the model's generalization capability on real-world data without fine-tuning, which further demonstrates its effectiveness in practical scenarios.

eess.AS↗

arXiv2025

Graph Transformers: A Survey

Graph transformers are a recent advancement in machine learning, offering a new class of neural network models for graph-structured data. The synergy between transformers and graph learning demonstrates strong performance and versatility across various graph-related tasks. This survey provides an in-depth review of recent progress and challenges in graph transformer research. We begin with foundational concepts of graphs and transformers. We then explore design perspectives of graph transformers, focusing on how they integrate graph inductive biases and graph attention mechanisms into the transformer architecture. Furthermore, we propose a taxonomy classifying graph transformers based on depth, scalability, and pre-training strategies, summarizing key principles for effective development of graph transformer models. Beyond technical analysis, we discuss the applications of graph transformer models for node-level, edge-level, and graph-level tasks, exploring their potential in other application scenarios as well. Finally, we identify remaining challenges in the field, such as scalability and efficiency, generalization and robustness, interpretability and explainability, dynamic and complex graphs, as well as data quality and diversity, charting future directions for graph transformer research.

cs.LG↗

arXiv2022

Inductive Graph Transformer for Delivery Time Estimation

Providing accurate estimated time of package delivery on users' purchasing pages for e-commerce platforms is of great importance to their purchasing decisions and post-purchase experiences. Although this problem shares some common issues with the conventional estimated time of arrival (ETA), it is more challenging with the following aspects: 1) Inductive inference. Models are required to predict ETA for orders with unseen retailers and addresses; 2) High-order interaction of order semantic information. Apart from the spatio-temporal features, the estimated time also varies greatly with other factors, such as the packaging efficiency of retailers, as well as the high-order interaction of these factors. In this paper, we propose an inductive graph transformer (IGT) that leverages raw feature information and structural graph data to estimate package delivery time. Different from previous graph transformer architectures, IGT adopts a decoupled pipeline and trains transformer as a regression function that can capture the multiplex information from both raw feature and dense embeddings encoded by a graph neural network (GNN). In addition, we further simplify the GNN structure by removing its non-linear activation and the learnable linear transformation matrix. The reduced parameter search space and linear information propagation in the simplified GNN enable the IGT to be applied in large-scale industrial scenarios. Experiments on real-world logistics datasets show that our proposed model can significantly outperform the state-of-the-art methods on estimation of delivery time. The source code is available at: https://github.com/enoche/IGT-WSDM23.

cs.LG↗

arXiv2025

Tricks and Plug-ins for Gradient Boosting with Transformers

Transformer architectures dominate modern NLP but often demand heavy computational resources and intricate hyperparameter tuning. To mitigate these challenges, we propose a novel framework, BoostTransformer, that augments transformers with boosting principles through subgrid token selection and importance-weighted sampling. Our method incorporates a least square boosting objective directly into the transformer pipeline, enabling more efficient training and improved performance. Across multiple fine-grained text classification benchmarks, BoostTransformer demonstrates both faster convergence and higher accuracy, surpassing standard transformers while minimizing architectural search overhead.

cs.LG↗

arXiv2025

Converting Transformers into DGNNs Form

Recent advances in deep learning have established Transformer architectures as the predominant modeling paradigm. Central to the success of Transformers is the self-attention mechanism, which scores the similarity between query and key matrices to modulate a value matrix. This operation bears striking similarities to digraph convolution, prompting an investigation into whether digraph convolution could serve as an alternative to self-attention. In this study, we formalize this concept by introducing a synthetic unitary digraph convolution based on the digraph Fourier transform. The resulting model, which we term Converter, effectively converts a Transformer into a Directed Graph Neural Network (DGNN) form. We have tested Converter on Long-Range Arena benchmark, long document classification, and DNA sequence-based taxonomy classification. Our experimental results demonstrate that Converter achieves superior performance while maintaining computational efficiency and architectural simplicity, which establishes it as a lightweight yet powerful Transformer variant.

cs.LG↗

arXiv2025

Tree Meets Transformer: A Hybrid Architecture for Scalable Power Allocation in Cell-Free Networks

Power allocation remains a fundamental challenge in wireless communication networks, particularly under dynamic user loads and large-scale deployments. While Transformerbased models have demonstrated strong performance, their computational cost scales poorly with the number of users. In this work, we propose a novel hybrid Tree-Transformer architecture that achieves scalable per-user power allocation. Our model compresses user features via a binary tree into a global root representation, applies a Transformer encoder solely to this root, and decodes per-user uplink and downlink powers through a shared decoder. This design achieves logarithmic depth and linear total complexity, enabling efficient inference across large and variable user sets without retraining or architectural changes. We evaluate our model on the max-min fairness problem in cellfree massive MIMO systems and demonstrate that it achieves near-optimal performance while significantly reducing inference time compared to full-attention baselines.

eess.SY↗

arXiv2025

Learnable Multi-Scale Wavelet Transformer: A Novel Alternative to Self-Attention

Transformer architectures, underpinned by the self-attention mechanism, have achieved state-of-the-art results across numerous natural language processing (NLP) tasks by effectively modeling long-range dependencies. However, the computational complexity of self-attention, scaling quadratically with input sequence length, presents significant challenges for processing very long sequences or operating under resource constraints. This paper introduces the Learnable Multi-Scale Wavelet Transformer (LMWT), a novel architecture that replaces the standard dot-product self-attention with a learnable multi-scale Haar wavelet transform module. Leveraging the intrinsic multi-resolution properties of wavelets, the LMWT efficiently captures both local details and global context. Crucially, the parameters of the wavelet transform, including scale-specific coefficients, are learned end-to-end during training, allowing the model to adapt its decomposition strategy to the data and task. We present the detailed mathematical formulation of the learnable Haar wavelet module and its integration into the transformer framework, supplemented by an architectural diagram. We conduct a comprehensive experimental evaluation on a standard machine translation benchmark (WMT16 En-De), comparing the LMWT against a baseline self-attention transformer using metrics like BLEU score, perplexity, and token accuracy. Furthermore, we analyze the computational complexity, highlighting the linear scaling of our approach, discuss its novelty in the context of related work, and explore the interpretability offered by visualizing the learned Haar coefficients. Our results indicate that the LMWT achieves competitive performance while offering substantial computational advantages, positioning it as a promising and novel alternative for efficient sequence modeling.

cs.LG↗

arXiv2026

Dissecting Representation Structure in Vision Transformers: A Rigorous Architectural Study

Representation structure is crucial for understanding Vision Transformer (ViT) architectures and their generalization behavior. However, prior studies neither isolate nor analyze module-level features nor investigate how their interactions contribute to performance estimation. In this work, we conduct the first rigorous analysis of feature information across diverse architectural scales, empirically uncover the relationship between ViT representation and generalization behavior, and leverage these insights to guide efficient ViT design. Our contributions are fivefold: Across diverse architectural scales, 1) We identify feature collapse at initialization, which leads to redundancy, and propose a reduction scheme to mitigate this issue. 2) We quantify feature information using entropy and the minimum eigenvalue, demonstrating that these metrics serve as reliable indicators for generalization prediction. 3) We show that feature in the token space provides a more faithful representation than those in embedding space. 4) We discover an unexpected finding: features produced by linear submodules within ViT layers are critical for the prediction of generalization performance. 5) Our proposed proxy improves the correlation ranking by 18-48% over prior baselines and can effectively identify ViT architectures that achieve higher accuracy at lower or comparable computational cost.

cs.CV↗

arXiv2021

Injecting Hierarchy with U-Net Transformers

The Transformer architecture has become increasingly popular over the past two years, owing to its impressive performance on a number of natural language processing (NLP) tasks. However, all Transformer computations occur at the level of word representations and therefore, it may be argued that Transformer models do not explicitly attempt to learn hierarchical structure which is widely assumed to be integral to language. In the present work, we introduce hierarchical processing into the Transformer model, taking inspiration from the U-Net architecture, popular in computer vision for its hierarchical view of natural images. We empirically demonstrate that the proposed architecture outperforms both the vanilla Transformer and some strong baselines in the domain of chit-chat dialogue.

cs.LG↗