{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:2SQZFW2DBOEQ47RXXZ3NIMTIGK","short_pith_number":"pith:2SQZFW2D","schema_version":"1.0","canonical_sha256":"d4a192db430b890e7e37be76d4326832a877e2527ef145865a24e4cc1435105f","source":{"kind":"arxiv","id":"2602.03024","version":2},"attestation_state":"computed","paper":{"title":"Consistency Deep Equilibrium Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Jingwen Xu, Junchao Lin, Robert C. Qiu, Zenan ling","submitted_at":"2026-02-03T02:42:48Z","abstract_excerpt":"Deep Equilibrium Models (DEQs) have emerged as a powerful paradigm in deep learning, offering the ability to model infinite-depth networks with constant memory usage. However, DEQs incur significant inference latency due to the iterative nature of fixed-point solvers. In this work, we introduce the Consistency Deep Equilibrium Model (C-DEQ), a novel framework that leverages consistency distillation to accelerate DEQ inference. We cast the DEQ iterative inference process as evolution along a fixed ODE trajectory toward the equilibrium. Along this trajectory, we train C-DEQs to consistently map "},"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":"2602.03024","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-02-03T02:42:48Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9f5213ba59fce2e4f0ab729a938d9e43503769af7051796016c28a5e1e73c94f","abstract_canon_sha256":"2cee030e45e508eafb1dc2041ca93de501143cd8775662e25a8290ab87ada554"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T02:04:13.804865Z","signature_b64":"kJcY91YM/JZf7kx9bJR4PVg5rWbF3jv/hbXLRlC65YDojMGbLp/UT8DtY0QZAbpcswvjBW5Rinc2BDop7NNTDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d4a192db430b890e7e37be76d4326832a877e2527ef145865a24e4cc1435105f","last_reissued_at":"2026-06-02T02:04:13.804330Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T02:04:13.804330Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Consistency Deep Equilibrium Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Jingwen Xu, Junchao Lin, Robert C. Qiu, Zenan ling","submitted_at":"2026-02-03T02:42:48Z","abstract_excerpt":"Deep Equilibrium Models (DEQs) have emerged as a powerful paradigm in deep learning, offering the ability to model infinite-depth networks with constant memory usage. However, DEQs incur significant inference latency due to the iterative nature of fixed-point solvers. In this work, we introduce the Consistency Deep Equilibrium Model (C-DEQ), a novel framework that leverages consistency distillation to accelerate DEQ inference. We cast the DEQ iterative inference process as evolution along a fixed ODE trajectory toward the equilibrium. Along this trajectory, we train C-DEQs to consistently map "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.03024","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/2602.03024/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":"2602.03024","created_at":"2026-06-02T02:04:13.804401+00:00"},{"alias_kind":"arxiv_version","alias_value":"2602.03024v2","created_at":"2026-06-02T02:04:13.804401+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.03024","created_at":"2026-06-02T02:04:13.804401+00:00"},{"alias_kind":"pith_short_12","alias_value":"2SQZFW2DBOEQ","created_at":"2026-06-02T02:04:13.804401+00:00"},{"alias_kind":"pith_short_16","alias_value":"2SQZFW2DBOEQ47RX","created_at":"2026-06-02T02:04:13.804401+00:00"},{"alias_kind":"pith_short_8","alias_value":"2SQZFW2D","created_at":"2026-06-02T02:04:13.804401+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.19705","citing_title":"i-DEQ: A stable inertial deep equilibrium model for image restoration","ref_index":62,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2SQZFW2DBOEQ47RXXZ3NIMTIGK","json":"https://pith.science/pith/2SQZFW2DBOEQ47RXXZ3NIMTIGK.json","graph_json":"https://pith.science/api/pith-number/2SQZFW2DBOEQ47RXXZ3NIMTIGK/graph.json","events_json":"https://pith.science/api/pith-number/2SQZFW2DBOEQ47RXXZ3NIMTIGK/events.json","paper":"https://pith.science/paper/2SQZFW2D"},"agent_actions":{"view_html":"https://pith.science/pith/2SQZFW2DBOEQ47RXXZ3NIMTIGK","download_json":"https://pith.science/pith/2SQZFW2DBOEQ47RXXZ3NIMTIGK.json","view_paper":"https://pith.science/paper/2SQZFW2D","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2602.03024&json=true","fetch_graph":"https://pith.science/api/pith-number/2SQZFW2DBOEQ47RXXZ3NIMTIGK/graph.json","fetch_events":"https://pith.science/api/pith-number/2SQZFW2DBOEQ47RXXZ3NIMTIGK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2SQZFW2DBOEQ47RXXZ3NIMTIGK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2SQZFW2DBOEQ47RXXZ3NIMTIGK/action/storage_attestation","attest_author":"https://pith.science/pith/2SQZFW2DBOEQ47RXXZ3NIMTIGK/action/author_attestation","sign_citation":"https://pith.science/pith/2SQZFW2DBOEQ47RXXZ3NIMTIGK/action/citation_signature","submit_replication":"https://pith.science/pith/2SQZFW2DBOEQ47RXXZ3NIMTIGK/action/replication_record"}},"created_at":"2026-06-02T02:04:13.804401+00:00","updated_at":"2026-06-02T02:04:13.804401+00:00"}