Database Paper Browser

Back to papers

A Memory Guided Transformer for Time Series Forecasting

Summary: Memformer: a Memory Guided Transformer that fuses patch-wise recurrent graph learning with global attention to model dynamic, time-varying correlations in long multivariate series. Uses an Alternating Memory Enhancer to reconcile local/global signals and robustly handle disrupted correlations, yielding SOTA long-term forecasts. (summarized by gpt-5-mini on Feb 09 2026)

Paper ID
13887
Venue
VLDB
Year
2025
Pagerank
-
Overall Rank
13,115 | 8.77%
DOI
10.14778/3705829.3705842

Incoming Non-self Citations Over Time

No non-self incoming citations found for this paper in this database.

Authors

Incoming Citations (Sorted by Pagerank)

Showing 3 of 3 citing papers.

Previous Page 1 / 1 Next

Outgoing Citations (Sorted by Pagerank)

Showing 10 of 10 cited papers.

Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.

Previous Page 1 / 1 Next

Semantically Similar Papers