{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/linear-complexity-randomized-self-attention","title":"Linear Complexity Randomized Self-attention Mechanism","arxiv_id":"2204.04667","date":"2022-04-10","proceeding":null,"authors":["Lin Zheng","Chong Wang","Lingpeng Kong"],"abstract":"Recently, random feature attentions (RFAs) are proposed to approximate the softmax attention in linear time and space complexity by linearizing the exponential kernel. In this paper, we first propose a novel perspective to understand the bias in such approximation by recasting RFAs as self-normalized importance samplers. This perspective further sheds light on an \\emph{unbiased} estimator for the whole softmax attention, called randomized attention (RA). RA constructs positive random features via query-specific distributions and enjoys greatly improved approximation fidelity, albeit exhibiting quadratic complexity. By combining the expressiveness in RA and the efficiency in RFA, we develop a novel linear complexity self-attention mechanism called linear randomized attention (LARA). Extensive experiments across various domains demonstrate that RA and LARA significantly improve the performance of RFAs by a substantial margin.","url_abs":"https://arxiv.org/abs/2204.04667v2","url_pdf":"https://arxiv.org/pdf/2204.04667v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"linear-complexity-randomized-self-attention","repo_url":"https://github.com/hkunlp/efficient-attention","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[],"methods":[{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2204.04667","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2204.04667"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/hkunlp/efficient-attention","reach":null}],"summary":{"ran":2,"ran_fixture":1,"ran_draft_wrong":1,"unverified":2},"by_repo_kind":{"listed":{"samples":6,"ran":4,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"23536350c4d837de","entry":"FlattenTranspose","repo":"hkunlp/efficient-attention","repo_kind":"listed","path":"efficient-attention/efficient_attention/lara.py","file_url":"https://github.com/hkunlp/efficient-attention/blob/HEAD/efficient-attention/efficient_attention/lara.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"23536350c4d837de"}},{"code_sha256_prefix":"bd857b15ac85b005","entry":"MultiheadAttention","repo":"hkunlp/efficient-attention","repo_kind":"listed","path":"efficient-attention/efficient_attention/lara.py","file_url":"https://github.com/hkunlp/efficient-attention/blob/HEAD/efficient-attention/efficient_attention/lara.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"bd857b15ac85b005"}},{"code_sha256_prefix":"0c94a009ab2a3631","entry":"prm_projection","repo":"hkunlp/efficient-attention","repo_kind":"listed","path":"efficient-attention/efficient_attention/lara.py","file_url":"https://github.com/hkunlp/efficient-attention/blob/HEAD/efficient-attention/efficient_attention/lara.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"0c94a009ab2a3631"}},{"code_sha256_prefix":"1fad32a65ad69c27","entry":"remove_prefix","repo":"hkunlp/efficient-attention","repo_kind":"listed","path":"efficient-attention/efficient_attention/lara.py","file_url":"https://github.com/hkunlp/efficient-attention/blob/HEAD/efficient-attention/efficient_attention/lara.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"1fad32a65ad69c27"}},{"code_sha256_prefix":"f1033e63692c7a57","entry":"LinearRA","repo":"hkunlp/efficient-attention","repo_kind":"listed","path":"efficient-attention/efficient_attention/lara.py","file_url":"https://github.com/hkunlp/efficient-attention/blob/HEAD/efficient-attention/efficient_attention/lara.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"f1033e63692c7a57"}},{"code_sha256_prefix":"42be9b05d3fbd3b9","entry":"add_nested_argument","repo":"hkunlp/efficient-attention","repo_kind":"listed","path":"efficient-attention/efficient_attention/lara.py","file_url":"https://github.com/hkunlp/efficient-attention/blob/HEAD/efficient-attention/efficient_attention/lara.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"42be9b05d3fbd3b9"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}