arXiv AI June 19, 2026 By Ashim Dhor, Rasel Mondal, Pin Yu Chen

ITNet: A Learnable Integral Transform That Subsumes Convolution, Attention, and Recurrence

arXiv:2606. 19538v1 Announce Type: new Abstract: Convolutional networks, recurrent networks, and transformers each encode different inductive biases -- locality, sequential memory, and content-dependent pairwise interaction -- and have remained mathematically distinct since their inception.

Summary generated by The Flow from the publisher's feed. The full article lives at arXiv AI.

llms

arXiv Machine Learning

Jun 3

Dynamic Short Convolutions Improve Transformers

arXiv:2606. 03825v1 Announce Type: new Abstract: Transformers have become the dominant architecture for large language models, largely due to the scalability and flexibility of attention, feed-forward layers, residual connections, and normalization.

By Oliver Sieberling, Bharat Runwal, Rameswar Panda, Yoon Kim

llms safety

arXiv Machine Learning

Jun 5

MSTN: A Lightweight and Fast Model for General TimeSeries Analysis

arXiv:2511. 20577v5 Announce Type: replace Abstract: Real-world time series often exhibit strong non-stationarity, complex nonlinear dynamics, and behavior expressed across multiple temporal scales, from rapid local fluctuations to slow-evolving long-range trends.

By Sumit S Shevtekar, Chandresh K Maurya

llms nlp benchmarks

arXiv AI

Jul 24

Scaling Interpretable Transformers with Parity Bottleneck Layers

arXiv:2607. 20652v1 Announce Type: cross Abstract: Language models are thought to exhibit the phenomenon of superposition, representing many more features than dimensions in their residual streams.

By Andrew Mack, Kraig Yuheng Tou, Mark Henry, Zhengxun Wu, Lauren Greenspan

llms efficiency safety

arXiv Machine Learning

Jun 5

Exact Linear Attention

arXiv:2605. 18848v3 Announce Type: replace Abstract: This paper introduces Exact Linear Attention (ELA), a mechanism that achieves linear computational complexity for Transformer attention by exploiting the exact decomposition property of kernel functions, thereby eliminating approximation error.

By Weinuo Ou

llms computer-vision reinforcement-learning efficiency safety

arXiv Machine Learning

Aug 4

UDT: Reconciling U-Nets and Diffusion Transformers with Data-Adaptive Token Reduction

arXiv:2608. 01298v1 Announce Type: cross Abstract: Diffusion Transformers (DiTs) have emerged as a core architecture in generative modeling due to their scalability and adaptability to multimodal tasks.

By Junno Yun, Ya\c{s}ar Utku Al\c{c}alar, Mehmet Ak\c{c}akaya

llms diffusion multimodal safety

arXiv AI

Jul 23

Structured Latent Space Modeling over Multi-Scale Temporal Patches for Multivariate Time Series Forecasting

arXiv:2607. 19404v1 Announce Type: cross Abstract: Multivariate time series encode structural patterns that unfold across multiple temporal scales, yet most forecasting backbones treat learned representations as transient byproducts of prediction, leaving the organizational geometry of these patterns underexploited.

By Xingsheng Chen, Deyu Yi, Siu-Ming Yiu

diffusion benchmarks safety

Related stories

Dynamic Short Convolutions Improve Transformers

MSTN: A Lightweight and Fast Model for General TimeSeries Analysis

Scaling Interpretable Transformers with Parity Bottleneck Layers

Exact Linear Attention

UDT: Reconciling U-Nets and Diffusion Transformers with Data-Adaptive Token Reduction

Structured Latent Space Modeling over Multi-Scale Temporal Patches for Multivariate Time Series Forecasting