{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:PIVD2U2A2GZBABJPBD7CP4QRAU","short_pith_number":"pith:PIVD2U2A","schema_version":"1.0","canonical_sha256":"7a2a3d5340d1b210052f08fe27f2110524bdf14fda70a7f002ab23a7c95c23bb","source":{"kind":"arxiv","id":"2410.11261","version":2},"attestation_state":"computed","paper":{"title":"Beyond Linear Approximations: A Novel Pruning Approach for Attention Matrix","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Jiangxuan Long, Yingyu Liang, Yufa Zhou, Zhao Song, Zhenmei Shi","submitted_at":"2024-10-15T04:35:56Z","abstract_excerpt":"Large Language Models (LLMs) have shown immense potential in enhancing various aspects of our daily lives, from conversational AI to search and AI assistants. However, their growing capabilities come at the cost of extremely large model sizes, making deployment on edge devices challenging due to memory and computational constraints. This paper introduces a novel approach to LLM weight pruning that directly optimizes for approximating the attention matrix, a core component of transformer architectures. Unlike existing methods that focus on linear approximations, our approach accounts for the no"},"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":"2410.11261","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-15T04:35:56Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"2b358e1885efbbe2a771630d8534f7cd208cea6fcd5f516c6bda66212120cb70","abstract_canon_sha256":"9f17efe88e5dc6d779a977ca814aec9cbe4a2e2212b72f8891e40d7f444e9d6f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:20:16.535969Z","signature_b64":"8xAsl7WxfKbBzAebf5tuRnzZcGNjzmRfIEIdVl2p2k6Gmr9NryopuYK3966DuQS8xdaAWbz4DT45YM3Xf5wkAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7a2a3d5340d1b210052f08fe27f2110524bdf14fda70a7f002ab23a7c95c23bb","last_reissued_at":"2026-07-05T10:20:16.535430Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:20:16.535430Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Beyond Linear Approximations: A Novel Pruning Approach for Attention Matrix","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Jiangxuan Long, Yingyu Liang, Yufa Zhou, Zhao Song, Zhenmei Shi","submitted_at":"2024-10-15T04:35:56Z","abstract_excerpt":"Large Language Models (LLMs) have shown immense potential in enhancing various aspects of our daily lives, from conversational AI to search and AI assistants. However, their growing capabilities come at the cost of extremely large model sizes, making deployment on edge devices challenging due to memory and computational constraints. This paper introduces a novel approach to LLM weight pruning that directly optimizes for approximating the attention matrix, a core component of transformer architectures. Unlike existing methods that focus on linear approximations, our approach accounts for the no"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.11261","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/2410.11261/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":"2410.11261","created_at":"2026-07-05T10:20:16.535494+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.11261v2","created_at":"2026-07-05T10:20:16.535494+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.11261","created_at":"2026-07-05T10:20:16.535494+00:00"},{"alias_kind":"pith_short_12","alias_value":"PIVD2U2A2GZB","created_at":"2026-07-05T10:20:16.535494+00:00"},{"alias_kind":"pith_short_16","alias_value":"PIVD2U2A2GZBABJP","created_at":"2026-07-05T10:20:16.535494+00:00"},{"alias_kind":"pith_short_8","alias_value":"PIVD2U2A","created_at":"2026-07-05T10:20:16.535494+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.18413","citing_title":"LatentLLM: Attention-Aware Joint Tensor Compression","ref_index":21,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PIVD2U2A2GZBABJPBD7CP4QRAU","json":"https://pith.science/pith/PIVD2U2A2GZBABJPBD7CP4QRAU.json","graph_json":"https://pith.science/api/pith-number/PIVD2U2A2GZBABJPBD7CP4QRAU/graph.json","events_json":"https://pith.science/api/pith-number/PIVD2U2A2GZBABJPBD7CP4QRAU/events.json","paper":"https://pith.science/paper/PIVD2U2A"},"agent_actions":{"view_html":"https://pith.science/pith/PIVD2U2A2GZBABJPBD7CP4QRAU","download_json":"https://pith.science/pith/PIVD2U2A2GZBABJPBD7CP4QRAU.json","view_paper":"https://pith.science/paper/PIVD2U2A","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.11261&json=true","fetch_graph":"https://pith.science/api/pith-number/PIVD2U2A2GZBABJPBD7CP4QRAU/graph.json","fetch_events":"https://pith.science/api/pith-number/PIVD2U2A2GZBABJPBD7CP4QRAU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PIVD2U2A2GZBABJPBD7CP4QRAU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PIVD2U2A2GZBABJPBD7CP4QRAU/action/storage_attestation","attest_author":"https://pith.science/pith/PIVD2U2A2GZBABJPBD7CP4QRAU/action/author_attestation","sign_citation":"https://pith.science/pith/PIVD2U2A2GZBABJPBD7CP4QRAU/action/citation_signature","submit_replication":"https://pith.science/pith/PIVD2U2A2GZBABJPBD7CP4QRAU/action/replication_record"}},"created_at":"2026-07-05T10:20:16.535494+00:00","updated_at":"2026-07-05T10:20:16.535494+00:00"}