{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:53UDTA6NHFGGBUS3EYHDQDCXIM","short_pith_number":"pith:53UDTA6N","schema_version":"1.0","canonical_sha256":"eee83983cd394c60d25b260e380c57431331db571164e030d07d2ca6da026c2a","source":{"kind":"arxiv","id":"2310.04064","version":1},"attestation_state":"computed","paper":{"title":"How to Capture Higher-order Correlations? Generalizing Matrix Softmax Attention to Kronecker Computation","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.CC","cs.CL","cs.LG","stat.ML"],"primary_cat":"cs.DS","authors_text":"Josh Alman, Zhao Song","submitted_at":"2023-10-06T07:42:39Z","abstract_excerpt":"In the classical transformer attention scheme, we are given three $n \\times d$ size matrices $Q, K, V$ (the query, key, and value tokens), and the goal is to compute a new $n \\times d$ size matrix $D^{-1} \\exp(QK^\\top) V$ where $D = \\mathrm{diag}( \\exp(QK^\\top) {\\bf 1}_n )$. In this work, we study a generalization of attention which captures triple-wise correlations. This generalization is able to solve problems about detecting triple-wise connections that were shown to be impossible for transformers. The potential downside of this generalization is that it appears as though computations are e"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2310.04064","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.DS","submitted_at":"2023-10-06T07:42:39Z","cross_cats_sorted":["cs.CC","cs.CL","cs.LG","stat.ML"],"title_canon_sha256":"3ed25a0e222618c79a70f82e75a81b0921b8ed889f1c1acb1c4fe9552b6e8c93","abstract_canon_sha256":"5ce5e7a3305b1fbbf1bff7bdc5c006779f145fd9d36815b0ca642f6a634d17db"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:57:57.827244Z","signature_b64":"kepdVP3ZFfEF359bmJ2EKTpHM7QwWJuf5I7J1ZjSc0TBCpsIIfSrWmBoAbKv5R2exaSdmfvi9b+HBFfLdHuZBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"eee83983cd394c60d25b260e380c57431331db571164e030d07d2ca6da026c2a","last_reissued_at":"2026-07-05T06:57:57.826752Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:57:57.826752Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"How to Capture Higher-order Correlations? Generalizing Matrix Softmax Attention to Kronecker Computation","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.CC","cs.CL","cs.LG","stat.ML"],"primary_cat":"cs.DS","authors_text":"Josh Alman, Zhao Song","submitted_at":"2023-10-06T07:42:39Z","abstract_excerpt":"In the classical transformer attention scheme, we are given three $n \\times d$ size matrices $Q, K, V$ (the query, key, and value tokens), and the goal is to compute a new $n \\times d$ size matrix $D^{-1} \\exp(QK^\\top) V$ where $D = \\mathrm{diag}( \\exp(QK^\\top) {\\bf 1}_n )$. In this work, we study a generalization of attention which captures triple-wise correlations. This generalization is able to solve problems about detecting triple-wise connections that were shown to be impossible for transformers. The potential downside of this generalization is that it appears as though computations are e"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.04064","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2310.04064/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2310.04064","created_at":"2026-07-05T06:57:57.826819+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.04064v1","created_at":"2026-07-05T06:57:57.826819+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.04064","created_at":"2026-07-05T06:57:57.826819+00:00"},{"alias_kind":"pith_short_12","alias_value":"53UDTA6NHFGG","created_at":"2026-07-05T06:57:57.826819+00:00"},{"alias_kind":"pith_short_16","alias_value":"53UDTA6NHFGGBUS3","created_at":"2026-07-05T06:57:57.826819+00:00"},{"alias_kind":"pith_short_8","alias_value":"53UDTA6N","created_at":"2026-07-05T06:57:57.826819+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2306.14048","citing_title":"H$_2$O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models","ref_index":118,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/53UDTA6NHFGGBUS3EYHDQDCXIM","json":"https://pith.science/pith/53UDTA6NHFGGBUS3EYHDQDCXIM.json","graph_json":"https://pith.science/api/pith-number/53UDTA6NHFGGBUS3EYHDQDCXIM/graph.json","events_json":"https://pith.science/api/pith-number/53UDTA6NHFGGBUS3EYHDQDCXIM/events.json","paper":"https://pith.science/paper/53UDTA6N"},"agent_actions":{"view_html":"https://pith.science/pith/53UDTA6NHFGGBUS3EYHDQDCXIM","download_json":"https://pith.science/pith/53UDTA6NHFGGBUS3EYHDQDCXIM.json","view_paper":"https://pith.science/paper/53UDTA6N","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.04064&json=true","fetch_graph":"https://pith.science/api/pith-number/53UDTA6NHFGGBUS3EYHDQDCXIM/graph.json","fetch_events":"https://pith.science/api/pith-number/53UDTA6NHFGGBUS3EYHDQDCXIM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/53UDTA6NHFGGBUS3EYHDQDCXIM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/53UDTA6NHFGGBUS3EYHDQDCXIM/action/storage_attestation","attest_author":"https://pith.science/pith/53UDTA6NHFGGBUS3EYHDQDCXIM/action/author_attestation","sign_citation":"https://pith.science/pith/53UDTA6NHFGGBUS3EYHDQDCXIM/action/citation_signature","submit_replication":"https://pith.science/pith/53UDTA6NHFGGBUS3EYHDQDCXIM/action/replication_record"}},"created_at":"2026-07-05T06:57:57.826819+00:00","updated_at":"2026-07-05T06:57:57.826819+00:00"}