{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:EFWPRL4GMSGA6PV6PXPJZBTUK2","short_pith_number":"pith:EFWPRL4G","schema_version":"1.0","canonical_sha256":"216cf8af86648c0f3ebe7dde9c867456b286636b3bc37d18255c60f8144399c4","source":{"kind":"arxiv","id":"2412.20875","version":1},"attestation_state":"computed","paper":{"title":"Attention Is All You Need For Mixture-of-Depths Routing","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Advait Gadhikar, Lukas Schott, Martin Rapp, Niclas Popp, Piyapat Saranrittichai, Souptik Kumar Majumdar","submitted_at":"2024-12-30T11:25:54Z","abstract_excerpt":"Advancements in deep learning are driven by training models with increasingly larger numbers of parameters, which in turn heightens the computational demands. To address this issue, Mixture-of-Depths (MoD) models have been proposed to dynamically assign computations only to the most relevant parts of the inputs, thereby enabling the deployment of large-parameter models with high efficiency during inference and training. These MoD models utilize a routing mechanism to determine which tokens should be processed by a layer, or skipped. However, conventional MoD models employ additional network la"},"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":"2412.20875","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.CV","submitted_at":"2024-12-30T11:25:54Z","cross_cats_sorted":[],"title_canon_sha256":"e2e07987c004609db2924b9c65db2d6425a6400983a6f236b843a34fd7e11ddb","abstract_canon_sha256":"5f3500a5bd65aad4683827fd614780228b11cca1ae83a4e181294b2adecbf1b3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:55:23.997036Z","signature_b64":"TjrgTY8fSUMSx4MrszUskeMDLPcwTJz9A1fnltXr50J/zUmoIssQnB+7I/OyE7sH49L+R92+loSVAYY/kvhQBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"216cf8af86648c0f3ebe7dde9c867456b286636b3bc37d18255c60f8144399c4","last_reissued_at":"2026-07-05T09:55:23.996536Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:55:23.996536Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Attention Is All You Need For Mixture-of-Depths Routing","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Advait Gadhikar, Lukas Schott, Martin Rapp, Niclas Popp, Piyapat Saranrittichai, Souptik Kumar Majumdar","submitted_at":"2024-12-30T11:25:54Z","abstract_excerpt":"Advancements in deep learning are driven by training models with increasingly larger numbers of parameters, which in turn heightens the computational demands. To address this issue, Mixture-of-Depths (MoD) models have been proposed to dynamically assign computations only to the most relevant parts of the inputs, thereby enabling the deployment of large-parameter models with high efficiency during inference and training. These MoD models utilize a routing mechanism to determine which tokens should be processed by a layer, or skipped. However, conventional MoD models employ additional network la"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.20875","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/2412.20875/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":"2412.20875","created_at":"2026-07-05T09:55:23.996595+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.20875v1","created_at":"2026-07-05T09:55:23.996595+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.20875","created_at":"2026-07-05T09:55:23.996595+00:00"},{"alias_kind":"pith_short_12","alias_value":"EFWPRL4GMSGA","created_at":"2026-07-05T09:55:23.996595+00:00"},{"alias_kind":"pith_short_16","alias_value":"EFWPRL4GMSGA6PV6","created_at":"2026-07-05T09:55:23.996595+00:00"},{"alias_kind":"pith_short_8","alias_value":"EFWPRL4G","created_at":"2026-07-05T09:55:23.996595+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EFWPRL4GMSGA6PV6PXPJZBTUK2","json":"https://pith.science/pith/EFWPRL4GMSGA6PV6PXPJZBTUK2.json","graph_json":"https://pith.science/api/pith-number/EFWPRL4GMSGA6PV6PXPJZBTUK2/graph.json","events_json":"https://pith.science/api/pith-number/EFWPRL4GMSGA6PV6PXPJZBTUK2/events.json","paper":"https://pith.science/paper/EFWPRL4G"},"agent_actions":{"view_html":"https://pith.science/pith/EFWPRL4GMSGA6PV6PXPJZBTUK2","download_json":"https://pith.science/pith/EFWPRL4GMSGA6PV6PXPJZBTUK2.json","view_paper":"https://pith.science/paper/EFWPRL4G","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.20875&json=true","fetch_graph":"https://pith.science/api/pith-number/EFWPRL4GMSGA6PV6PXPJZBTUK2/graph.json","fetch_events":"https://pith.science/api/pith-number/EFWPRL4GMSGA6PV6PXPJZBTUK2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EFWPRL4GMSGA6PV6PXPJZBTUK2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EFWPRL4GMSGA6PV6PXPJZBTUK2/action/storage_attestation","attest_author":"https://pith.science/pith/EFWPRL4GMSGA6PV6PXPJZBTUK2/action/author_attestation","sign_citation":"https://pith.science/pith/EFWPRL4GMSGA6PV6PXPJZBTUK2/action/citation_signature","submit_replication":"https://pith.science/pith/EFWPRL4GMSGA6PV6PXPJZBTUK2/action/replication_record"}},"created_at":"2026-07-05T09:55:23.996595+00:00","updated_at":"2026-07-05T09:55:23.996595+00:00"}