{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:B7A2URNVWKOESGJGBFHPTJCDO2","short_pith_number":"pith:B7A2URNV","schema_version":"1.0","canonical_sha256":"0fc1aa45b5b29c491926094ef9a4437686e90243176cf424991b09beacd9958f","source":{"kind":"arxiv","id":"2504.21463","version":2},"attestation_state":"computed","paper":{"title":"RWKV-X: A Linear Complexity Hybrid Language Model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Fei Richard Yu, Haowen Hou, Kaifeng Tan, Rongchang Lu, Zhiyi Huang","submitted_at":"2025-04-30T09:38:17Z","abstract_excerpt":"In this paper, we introduce RWKV-X, a novel hybrid architecture that combines the efficiency of RWKV for short-range modeling with a sparse attention mechanism designed to capture long-range context. Unlike previous hybrid approaches that rely on full attention layers and retain quadratic complexity, RWKV-X achieves linear-time complexity in training and constant-time complexity in inference decoding. We demonstrate that RWKV-X, when continually pretrained on 64K-token sequences, achieves near-perfect accuracy on the 64K passkey retrieval benchmark. It consistently outperforms prior RWKV-7 mod"},"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":"2504.21463","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-30T09:38:17Z","cross_cats_sorted":[],"title_canon_sha256":"5005e55a0421a6b604946928759f0b3f6fc1a9902138db6b5070a16e8b257851","abstract_canon_sha256":"8d2c014b7496f09a20b309a60045589fd7a7d346f7b395281959fb2b8c6f8900"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:00:44.354996Z","signature_b64":"IzRco6qn7W5LPJVEzfaqW5wCHkmp2jScp0cnXCUa3y1WeUTXqcsrCFQHUztRBbwYqvgQwpj4ySc6tMn1Jg+wAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0fc1aa45b5b29c491926094ef9a4437686e90243176cf424991b09beacd9958f","last_reissued_at":"2026-07-05T11:00:44.354515Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:00:44.354515Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RWKV-X: A Linear Complexity Hybrid Language Model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Fei Richard Yu, Haowen Hou, Kaifeng Tan, Rongchang Lu, Zhiyi Huang","submitted_at":"2025-04-30T09:38:17Z","abstract_excerpt":"In this paper, we introduce RWKV-X, a novel hybrid architecture that combines the efficiency of RWKV for short-range modeling with a sparse attention mechanism designed to capture long-range context. Unlike previous hybrid approaches that rely on full attention layers and retain quadratic complexity, RWKV-X achieves linear-time complexity in training and constant-time complexity in inference decoding. We demonstrate that RWKV-X, when continually pretrained on 64K-token sequences, achieves near-perfect accuracy on the 64K passkey retrieval benchmark. It consistently outperforms prior RWKV-7 mod"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.21463","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/2504.21463/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":"2504.21463","created_at":"2026-07-05T11:00:44.354572+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.21463v2","created_at":"2026-07-05T11:00:44.354572+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.21463","created_at":"2026-07-05T11:00:44.354572+00:00"},{"alias_kind":"pith_short_12","alias_value":"B7A2URNVWKOE","created_at":"2026-07-05T11:00:44.354572+00:00"},{"alias_kind":"pith_short_16","alias_value":"B7A2URNVWKOESGJG","created_at":"2026-07-05T11:00:44.354572+00:00"},{"alias_kind":"pith_short_8","alias_value":"B7A2URNV","created_at":"2026-07-05T11:00:44.354572+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.14926","citing_title":"SCRWKV: Ultra-Compact Structure-Calibrated Vision-RWKV for Topological Crack Segmentation","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2510.04800","citing_title":"Hybrid Architectures for Language Models: Systematic Analysis and Design Insights","ref_index":22,"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":15,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/B7A2URNVWKOESGJGBFHPTJCDO2","json":"https://pith.science/pith/B7A2URNVWKOESGJGBFHPTJCDO2.json","graph_json":"https://pith.science/api/pith-number/B7A2URNVWKOESGJGBFHPTJCDO2/graph.json","events_json":"https://pith.science/api/pith-number/B7A2URNVWKOESGJGBFHPTJCDO2/events.json","paper":"https://pith.science/paper/B7A2URNV"},"agent_actions":{"view_html":"https://pith.science/pith/B7A2URNVWKOESGJGBFHPTJCDO2","download_json":"https://pith.science/pith/B7A2URNVWKOESGJGBFHPTJCDO2.json","view_paper":"https://pith.science/paper/B7A2URNV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.21463&json=true","fetch_graph":"https://pith.science/api/pith-number/B7A2URNVWKOESGJGBFHPTJCDO2/graph.json","fetch_events":"https://pith.science/api/pith-number/B7A2URNVWKOESGJGBFHPTJCDO2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/B7A2URNVWKOESGJGBFHPTJCDO2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/B7A2URNVWKOESGJGBFHPTJCDO2/action/storage_attestation","attest_author":"https://pith.science/pith/B7A2URNVWKOESGJGBFHPTJCDO2/action/author_attestation","sign_citation":"https://pith.science/pith/B7A2URNVWKOESGJGBFHPTJCDO2/action/citation_signature","submit_replication":"https://pith.science/pith/B7A2URNVWKOESGJGBFHPTJCDO2/action/replication_record"}},"created_at":"2026-07-05T11:00:44.354572+00:00","updated_at":"2026-07-05T11:00:44.354572+00:00"}