{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:VTKT73LJLVKIL3VRCE2MZCKXDD","short_pith_number":"pith:VTKT73LJ","schema_version":"1.0","canonical_sha256":"acd53fed695d5485eeb11134cc895718d350d1e94d0e4a9ff983f522b2724c97","source":{"kind":"arxiv","id":"2405.06640","version":1},"attestation_state":"computed","paper":{"title":"Linearizing Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Achal Dave, Adrien Gaidon, Igor Vasiljevic, Jean Mercat, Kushal Arora, Sedrick Keh, Thomas Kollar","submitted_at":"2024-05-10T17:59:08Z","abstract_excerpt":"Linear transformers have emerged as a subquadratic-time alternative to softmax attention and have garnered significant interest due to their fixed-size recurrent state that lowers inference cost. However, their original formulation suffers from poor scaling and underperforms compute-matched transformers. Recent linear models such as RWKV and Mamba have attempted to address these shortcomings by proposing novel time-mixing and gating architectures, but pre-training large language models requires significant data and compute investments. Thus, the search for subquadratic architectures is limited"},"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":"2405.06640","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-10T17:59:08Z","cross_cats_sorted":[],"title_canon_sha256":"38315448a961913d76cbe6e19fb223d80317ef6f2db20b8a89c20c7e51f428a9","abstract_canon_sha256":"38184110be4b29ed953ab3027fdd6a3dab93cc5bf4f0bed45c178c7e0d3f1d9a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:17:52.603329Z","signature_b64":"N5YEEj+1sh8LpXIxiu2o9ctBd0fudfQq1oWhMBnUtmtBlczXxnOH+EBJGgg8ujTrXgT8njCQLphe1Q34JXDkBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"acd53fed695d5485eeb11134cc895718d350d1e94d0e4a9ff983f522b2724c97","last_reissued_at":"2026-07-05T08:17:52.602840Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:17:52.602840Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Linearizing Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Achal Dave, Adrien Gaidon, Igor Vasiljevic, Jean Mercat, Kushal Arora, Sedrick Keh, Thomas Kollar","submitted_at":"2024-05-10T17:59:08Z","abstract_excerpt":"Linear transformers have emerged as a subquadratic-time alternative to softmax attention and have garnered significant interest due to their fixed-size recurrent state that lowers inference cost. However, their original formulation suffers from poor scaling and underperforms compute-matched transformers. Recent linear models such as RWKV and Mamba have attempted to address these shortcomings by proposing novel time-mixing and gating architectures, but pre-training large language models requires significant data and compute investments. Thus, the search for subquadratic architectures is limited"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.06640","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/2405.06640/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":"2405.06640","created_at":"2026-07-05T08:17:52.602898+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.06640v1","created_at":"2026-07-05T08:17:52.602898+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.06640","created_at":"2026-07-05T08:17:52.602898+00:00"},{"alias_kind":"pith_short_12","alias_value":"VTKT73LJLVKI","created_at":"2026-07-05T08:17:52.602898+00:00"},{"alias_kind":"pith_short_16","alias_value":"VTKT73LJLVKIL3VR","created_at":"2026-07-05T08:17:52.602898+00:00"},{"alias_kind":"pith_short_8","alias_value":"VTKT73LJ","created_at":"2026-07-05T08:17:52.602898+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":9,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.07706","citing_title":"The Key to Going Linear: Analysis-Driven Transformer Linearization","ref_index":13,"is_internal_anchor":true},{"citing_arxiv_id":"2606.30562","citing_title":"Morphing into Hybrid Attention Models","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2605.25676","citing_title":"Llamion Technical Report","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2410.13846","citing_title":"LightTransfer: Your Long-Context LLM is Secretly a Hybrid Model with Effortless Adaptation","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2507.09025","citing_title":"Lizard: An Efficient Linearization Framework for Large Language Models","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2509.05276","citing_title":"SpikingBrain: Spiking Brain-inspired Large Models","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2502.13189","citing_title":"MoBA: Mixture of Block Attention for Long-Context LLMs","ref_index":52,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14191","citing_title":"Attention to Mamba: A Recipe for Cross-Architecture Distillation","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2604.22575","citing_title":"SpikingBrain2.0: Brain-Inspired Foundation Models for Efficient Long-Context and Cross-Platform Inference","ref_index":26,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VTKT73LJLVKIL3VRCE2MZCKXDD","json":"https://pith.science/pith/VTKT73LJLVKIL3VRCE2MZCKXDD.json","graph_json":"https://pith.science/api/pith-number/VTKT73LJLVKIL3VRCE2MZCKXDD/graph.json","events_json":"https://pith.science/api/pith-number/VTKT73LJLVKIL3VRCE2MZCKXDD/events.json","paper":"https://pith.science/paper/VTKT73LJ"},"agent_actions":{"view_html":"https://pith.science/pith/VTKT73LJLVKIL3VRCE2MZCKXDD","download_json":"https://pith.science/pith/VTKT73LJLVKIL3VRCE2MZCKXDD.json","view_paper":"https://pith.science/paper/VTKT73LJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.06640&json=true","fetch_graph":"https://pith.science/api/pith-number/VTKT73LJLVKIL3VRCE2MZCKXDD/graph.json","fetch_events":"https://pith.science/api/pith-number/VTKT73LJLVKIL3VRCE2MZCKXDD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VTKT73LJLVKIL3VRCE2MZCKXDD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VTKT73LJLVKIL3VRCE2MZCKXDD/action/storage_attestation","attest_author":"https://pith.science/pith/VTKT73LJLVKIL3VRCE2MZCKXDD/action/author_attestation","sign_citation":"https://pith.science/pith/VTKT73LJLVKIL3VRCE2MZCKXDD/action/citation_signature","submit_replication":"https://pith.science/pith/VTKT73LJLVKIL3VRCE2MZCKXDD/action/replication_record"}},"created_at":"2026-07-05T08:17:52.602898+00:00","updated_at":"2026-07-05T08:17:52.602898+00:00"}