Papers › LSG Attention: Extrapolation of pretrained Transformers to long sequences

LSG Attention: Extrapolation of pretrained Transformers to long sequences

13 Oct 2022arXiv:2210.15497archive 2025-07-28

Charles Condevaux, Sébastien Harispe

Transformer models achieve state-of-the-art performance on a wide range of NLP tasks. They however suffer from a prohibitive limitation due to the self-attention mechanism, inducing O(n²) complexity with regard to sequence length. To answer this limitation we introduce the LSG architecture which relies on Local, Sparse and Global attention. We show that LSG attention is fast, efficient and competitive in classification and summarization tasks on long documents. Interestingly, it can also be used to adapt existing pretrained models to efficiently extrapolate to longer sequences with no additional training. Along with the introduction of the LSG attention mechanism, we propose tools to train new models and adapt existing ones based on this mechanism.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

ccdv-ai/convert_checkpoint_to_lsg officialmentioned in papermentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections