Flexformer learns attention kernels by treating spectral frequencies as trainable parameters in random Fourier feature-based linear attention, with stationary and nonstationary variants that outperform fixed-kernel baselines.
itransformer: Inverted transformers are effective for time series forecasting
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 2years
2026 2verdicts
UNVERDICTED 2roles
background 1polarities
background 1representative citing papers
TOA augments attention with learnable sequence-space operators and stochastic regularization to enable signed temporal mixing, yielding gains on forecasting and related benchmarks when added to PatchTST and iTransformer.
citing papers explorer
-
Flexformer: Flexible Linear Transformer with Learnable Attention Kernel
Flexformer learns attention kernels by treating spectral frequencies as trainable parameters in random Fourier feature-based linear attention, with stationary and nonstationary variants that outperform fixed-kernel baselines.
-
Beyond Similarity: Temporal Operator Attention for Time Series Analysis
TOA augments attention with learnable sequence-space operators and stochastic regularization to enable signed temporal mixing, yielding gains on forecasting and related benchmarks when added to PatchTST and iTransformer.