Rotary Masked Autoencoders are Versatile Learners

• URL: http://arxiv.org/abs/2505.20535v1
• Date: Mon, 26 May 2025 21:45:18 GMT
• Title: Rotary Masked Autoencoders are Versatile Learners
• Authors: Uros Zivanovic, Serafina Di Gioia, Andre Scaffidi, Martín de los Rios, Gabriella Contardo, Roberto Trotta,
• Abstract summary: We present the Rotary Masked Autoencoder (RoMAE)<n>RoMAE is an extension to the Masked Autoencoder (MAE) that enables representation learning with multidimensional continuous positional information.<n>We showcase RoMAE's performance on a variety of modalities including irregular and multivariate time-series, images, and audio.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Applying Transformers to irregular time-series typically requires specializations to their baseline architecture, which can result in additional computational overhead and increased method complexity. We present the Rotary Masked Autoencoder (RoMAE), which utilizes the popular Rotary Positional Embedding (RoPE) method for continuous positions. RoMAE is an extension to the Masked Autoencoder (MAE) that enables representation learning with multidimensional continuous positional information while avoiding any time-series-specific architectural specializations. We showcase RoMAE's performance on a variety of modalities including irregular and multivariate time-series, images, and audio, demonstrating that RoMAE surpasses specialized time-series architectures on difficult datasets such as the DESC ELAsTiCC Challenge while maintaining MAE's usual performance across other modalities. In addition, we investigate RoMAE's ability to reconstruct the embedded continuous positions, demonstrating that including learned embeddings in the input sequence breaks RoPE's relative position property.

Related papers

• Rotary Outliers and Rotary Offset Features in Large Language Models [1.9580473532948401]
  We study the features and patterns that emerge in queries and keys when using rotary embeddings.<n>We find and analyze outliers across models in queries and keys and find that they are likely to be found in rotary features with partial cycles.
  arXiv  Detail & Related papers  (2025-03-03T18:55:09Z)
• VideoRoPE: What Makes for Good Video Rotary Position Embedding? [109.88966080843608]
  VideoRoPE consistently surpasses previous RoPE variants, across diverse downstream tasks such as long video retrieval, video understanding, and video hallucination.<n>VideoRoPE features textlow-frequency temporal allocation to mitigate periodic oscillations, a textitdiagonal layout to maintain spatial symmetry, and textadjustable temporal spacing to decouple temporal and spatial indexing.
  arXiv  Detail & Related papers  (2025-02-07T18:56:04Z)
• WaveRoRA: Wavelet Rotary Route Attention for Multivariate Time Series Forecasting [4.680374146155483]
  We propose a wavelet learning framework to model complex temporal dependencies of the time series data. The wavelet domain integrates both time and frequency information, allowing for the analysis of local characteristics of signals at different scales. We propose a novel attention mechanism: Rotary Route Attention (RoRA)
  arXiv  Detail & Related papers  (2024-10-30T02:36:55Z)
• PRformer: Pyramidal Recurrent Transformer for Multivariate Time Series Forecasting [82.03373838627606]
  Self-attention mechanism in Transformer architecture requires positional embeddings to encode temporal order in time series prediction. We argue that this reliance on positional embeddings restricts the Transformer's ability to effectively represent temporal sequences. We present a model integrating PRE with a standard Transformer encoder, demonstrating state-of-the-art performance on various real-world datasets.
  arXiv  Detail & Related papers  (2024-08-20T01:56:07Z)
• A Temporal Kolmogorov-Arnold Transformer for Time Series Forecasting [0.0]
  Temporal Kolmogorov-Arnold Transformer (TKAT) is a novel attention-based architecture designed to capture temporal patterns in data streams. Inspired by the Temporal Fusion Transformer (TFT), TKAT emerges as a powerful encoder-decoder model tailored to handle tasks in which the observed part of the features is more important than the a priori known part.
  arXiv  Detail & Related papers  (2024-06-04T16:55:42Z)
• UMMAFormer: A Universal Multimodal-adaptive Transformer Framework for Temporal Forgery Localization [16.963092523737593]
  We propose a novel framework for temporal forgery localization (TFL) that predicts forgery segments with multimodal adaptation. Our approach achieves state-of-the-art performance on benchmark datasets, including Lav-DF, TVIL, and Psynd.
  arXiv  Detail & Related papers  (2023-08-28T08:20:30Z)
• TimeMAE: Self-Supervised Representations of Time Series with Decoupled Masked Autoencoders [55.00904795497786]
  We propose TimeMAE, a novel self-supervised paradigm for learning transferrable time series representations based on transformer networks. The TimeMAE learns enriched contextual representations of time series with a bidirectional encoding scheme. To solve the discrepancy issue incurred by newly injected masked embeddings, we design a decoupled autoencoder architecture.
  arXiv  Detail & Related papers  (2023-03-01T08:33:16Z)
• FormerTime: Hierarchical Multi-Scale Representations for Multivariate Time Series Classification [53.55504611255664]
  FormerTime is a hierarchical representation model for improving the classification capacity for the multivariate time series classification task. It exhibits three aspects of merits: (1) learning hierarchical multi-scale representations from time series data, (2) inheriting the strength of both transformers and convolutional networks, and (3) tacking the efficiency challenges incurred by the self-attention mechanism.
  arXiv  Detail & Related papers  (2023-02-20T07:46:14Z)
• SatMAE: Pre-training Transformers for Temporal and Multi-Spectral Satellite Imagery [74.82821342249039]
  We present SatMAE, a pre-training framework for temporal or multi-spectral satellite imagery based on Masked Autoencoder (MAE) To leverage temporal information, we include a temporal embedding along with independently masking image patches across time.
  arXiv  Detail & Related papers  (2022-07-17T01:35:29Z)
• Improving Video Instance Segmentation via Temporal Pyramid Routing [61.10753640148878]
  Video Instance (VIS) is a new and inherently multi-task problem, which aims to detect, segment and track each instance in a video sequence. We propose a Temporal Pyramid Routing (TPR) strategy to conditionally align and conduct pixel-level aggregation from a feature pyramid pair of two adjacent frames. Our approach is a plug-and-play module and can be easily applied to existing instance segmentation methods.
  arXiv  Detail & Related papers  (2021-07-28T03:57:12Z)
• RoFormer: Enhanced Transformer with Rotary Position Embedding [9.01819510933327]
  We propose a novel method named Rotary Position Embedding(RoPE) to effectively leverage the positional information. RoPE encodes the absolute position with a rotation matrix and meanwhile incorporates the explicit relative position dependency in self-attention formulation. We evaluate the enhanced transformer with rotary position embedding, also called RoFormer, on various long text classification benchmark datasets.
  arXiv  Detail & Related papers  (2021-04-20T09:54:06Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.