{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:S7WKVTISDVBV3GEQD4VLMWC5AZ","short_pith_number":"pith:S7WKVTIS","schema_version":"1.0","canonical_sha256":"97ecaacd121d435d98901f2ab6585d06754fe55063aa1d70ef357a6cd1cca6e5","source":{"kind":"arxiv","id":"2507.20096","version":2},"attestation_state":"computed","paper":{"title":"EcoTransformer: Attention without Multiplication","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Hong Xu, Shirin Amiraslani, Xin Gao, Xingming Xu","submitted_at":"2025-07-27T01:32:54Z","abstract_excerpt":"The Transformer, with its scaled dot-product attention mechanism, has become a foundational architecture in modern AI. However, this mechanism is computationally intensive and incurs substantial energy costs. We propose a new Transformer architecture EcoTransformer, in which the output context vector is constructed as the convolution of the values using a Laplacian kernel, where the distances are measured by the L1 metric between the queries and keys. Compared to dot-product based attention, the new attention score calculation is free of matrix multiplication. It performs on par with, or even "},"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":"2507.20096","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-07-27T01:32:54Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"2fe3ff612f0e8500243839eb901a487f47880f0b4c1ca96d10d56554012a53b4","abstract_canon_sha256":"e0127f9d94f4d3f36edb0e610c1f5d7dd53be07acccd20cbdbaa9c89f1e42a40"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:49:11.857376Z","signature_b64":"7sgEDNdf+h7inKBy/lRgdiju+3UyHXK1Csv48ytPSiDUqWMSXhn4BM2ab+glmpTObFJZkozH1IOrA4KzDHWPBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"97ecaacd121d435d98901f2ab6585d06754fe55063aa1d70ef357a6cd1cca6e5","last_reissued_at":"2026-07-05T11:49:11.856896Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:49:11.856896Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"EcoTransformer: Attention without Multiplication","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Hong Xu, Shirin Amiraslani, Xin Gao, Xingming Xu","submitted_at":"2025-07-27T01:32:54Z","abstract_excerpt":"The Transformer, with its scaled dot-product attention mechanism, has become a foundational architecture in modern AI. However, this mechanism is computationally intensive and incurs substantial energy costs. We propose a new Transformer architecture EcoTransformer, in which the output context vector is constructed as the convolution of the values using a Laplacian kernel, where the distances are measured by the L1 metric between the queries and keys. Compared to dot-product based attention, the new attention score calculation is free of matrix multiplication. It performs on par with, or even "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.20096","kind":"arxiv","version":2},"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/2507.20096/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":"2507.20096","created_at":"2026-07-05T11:49:11.856953+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.20096v2","created_at":"2026-07-05T11:49:11.856953+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.20096","created_at":"2026-07-05T11:49:11.856953+00:00"},{"alias_kind":"pith_short_12","alias_value":"S7WKVTISDVBV","created_at":"2026-07-05T11:49:11.856953+00:00"},{"alias_kind":"pith_short_16","alias_value":"S7WKVTISDVBV3GEQ","created_at":"2026-07-05T11:49:11.856953+00:00"},{"alias_kind":"pith_short_8","alias_value":"S7WKVTIS","created_at":"2026-07-05T11:49:11.856953+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.21848","citing_title":"Keyless Attention: Value-Space Routing and Value-Only Caching for Efficient Transformers","ref_index":72,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/S7WKVTISDVBV3GEQD4VLMWC5AZ","json":"https://pith.science/pith/S7WKVTISDVBV3GEQD4VLMWC5AZ.json","graph_json":"https://pith.science/api/pith-number/S7WKVTISDVBV3GEQD4VLMWC5AZ/graph.json","events_json":"https://pith.science/api/pith-number/S7WKVTISDVBV3GEQD4VLMWC5AZ/events.json","paper":"https://pith.science/paper/S7WKVTIS"},"agent_actions":{"view_html":"https://pith.science/pith/S7WKVTISDVBV3GEQD4VLMWC5AZ","download_json":"https://pith.science/pith/S7WKVTISDVBV3GEQD4VLMWC5AZ.json","view_paper":"https://pith.science/paper/S7WKVTIS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.20096&json=true","fetch_graph":"https://pith.science/api/pith-number/S7WKVTISDVBV3GEQD4VLMWC5AZ/graph.json","fetch_events":"https://pith.science/api/pith-number/S7WKVTISDVBV3GEQD4VLMWC5AZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/S7WKVTISDVBV3GEQD4VLMWC5AZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/S7WKVTISDVBV3GEQD4VLMWC5AZ/action/storage_attestation","attest_author":"https://pith.science/pith/S7WKVTISDVBV3GEQD4VLMWC5AZ/action/author_attestation","sign_citation":"https://pith.science/pith/S7WKVTISDVBV3GEQD4VLMWC5AZ/action/citation_signature","submit_replication":"https://pith.science/pith/S7WKVTISDVBV3GEQD4VLMWC5AZ/action/replication_record"}},"created_at":"2026-07-05T11:49:11.856953+00:00","updated_at":"2026-07-05T11:49:11.856953+00:00"}