Long Context & Efficient Sequences · 2020

Sparse Attention (Longformer / BigBird)

Iz Beltagy, Matthew E. Peters, Arman Cohan, Manzil Zaheer, Amr Ahmed

Sparse-attention transformers cut the quadratic cost of full attention to linear in sequence length by attending over a fixed sparse pattern, with BigBird combining local, global, and random attention while proving the pattern still preserves the transformer's expressive and theoretical properties.

Editorial record

Plain-language summary

Full self-attention costs grow with the square of sequence length, capping practical context. Longformer uses sliding-window local attention plus a few global tokens; BigBird adds random connections and shows this local+global+random combination is a universal approximator and Turing-complete, so the sparsity does not sacrifice power. By making attention scale linearly, these models pushed usable context from around 512 into the thousands of tokens for tasks like long-document QA and genomics.

Knowledge graph

Relationships

Descendants

Source record

Provenance

Record ID
P-420
Record created
2026-07-13
Last reviewed
2026-07-14
Record version
2

Citation caveat: Citation metadata is approximate and marked unverified in the source dataset.