{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:G67C5S42GUMHL7DAE5ULLATSNJ","short_pith_number":"pith:G67C5S42","schema_version":"1.0","canonical_sha256":"37be2ecb9a351875fc602768b582726a6f84ad891744c56fd0e9cb7d690c6bf0","source":{"kind":"arxiv","id":"2608.01630","version":1},"attestation_state":"computed","paper":{"title":"RING: Retrieval-Internalized Generation for Continual Large-Scale Knowledge Injection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Huawei Shen, Jingcheng Deng, Liang Pang, Liyi Chen, Shicheng Xu, Xueqi Cheng, Yan Gao, Yao Hu, Yi Wu, Zihao Wei","submitted_at":"2026-08-03T03:00:43Z","abstract_excerpt":"Retrieval-augmented generation (RAG) improves factuality but adds latency and engineering overhead at serving time. We propose RING (Retrieval-Internalized Generation), a holistic paradigm spanning both architecture and training that injects large-scale external knowledge into a \\textit{Mixture-of-Memory Experts} and learns parametric search over this internal memory via reinforcement learning, removing the external retriever entirely. Training proceeds in three stages: continued pre-training injects new corpora into a Knowledge Expert via our novel \\textit{Dual Causal Attention}; supervised f"},"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":"2608.01630","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-08-03T03:00:43Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b02d31a96c9432bfe737e21b3a4fffcb04b1c7867f1188993df9517b396ba73a","abstract_canon_sha256":"8f4c09b0a59744e0715d43e8bd9f7dfa6167d69733ac7cf003680ea5187d298d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-04T02:06:09.865299Z","signature_b64":"siUppmUl/s8QzR06hdE9/xUTiI0V8k7PNxLDuaoKb7Q0gYWX/6a+kLk45ePtuLF96Q+UPjKi11tR3kC97GFLBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"37be2ecb9a351875fc602768b582726a6f84ad891744c56fd0e9cb7d690c6bf0","last_reissued_at":"2026-08-04T02:06:09.863878Z","signature_status":"signed_v1","first_computed_at":"2026-08-04T02:06:09.863878Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RING: Retrieval-Internalized Generation for Continual Large-Scale Knowledge Injection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Huawei Shen, Jingcheng Deng, Liang Pang, Liyi Chen, Shicheng Xu, Xueqi Cheng, Yan Gao, Yao Hu, Yi Wu, Zihao Wei","submitted_at":"2026-08-03T03:00:43Z","abstract_excerpt":"Retrieval-augmented generation (RAG) improves factuality but adds latency and engineering overhead at serving time. We propose RING (Retrieval-Internalized Generation), a holistic paradigm spanning both architecture and training that injects large-scale external knowledge into a \\textit{Mixture-of-Memory Experts} and learns parametric search over this internal memory via reinforcement learning, removing the external retriever entirely. Training proceeds in three stages: continued pre-training injects new corpora into a Knowledge Expert via our novel \\textit{Dual Causal Attention}; supervised f"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.01630","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/2608.01630/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":"2608.01630","created_at":"2026-08-04T02:06:09.865282+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.01630v1","created_at":"2026-08-04T02:06:09.865282+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.01630","created_at":"2026-08-04T02:06:09.865282+00:00"},{"alias_kind":"pith_short_12","alias_value":"G67C5S42GUMH","created_at":"2026-08-04T02:06:09.865282+00:00"},{"alias_kind":"pith_short_16","alias_value":"G67C5S42GUMHL7DA","created_at":"2026-08-04T02:06:09.865282+00:00"},{"alias_kind":"pith_short_8","alias_value":"G67C5S42","created_at":"2026-08-04T02:06:09.865282+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/G67C5S42GUMHL7DAE5ULLATSNJ","json":"https://pith.science/pith/G67C5S42GUMHL7DAE5ULLATSNJ.json","graph_json":"https://pith.science/api/pith-number/G67C5S42GUMHL7DAE5ULLATSNJ/graph.json","events_json":"https://pith.science/api/pith-number/G67C5S42GUMHL7DAE5ULLATSNJ/events.json","paper":"https://pith.science/paper/G67C5S42"},"agent_actions":{"view_html":"https://pith.science/pith/G67C5S42GUMHL7DAE5ULLATSNJ","download_json":"https://pith.science/pith/G67C5S42GUMHL7DAE5ULLATSNJ.json","view_paper":"https://pith.science/paper/G67C5S42","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.01630&json=true","fetch_graph":"https://pith.science/api/pith-number/G67C5S42GUMHL7DAE5ULLATSNJ/graph.json","fetch_events":"https://pith.science/api/pith-number/G67C5S42GUMHL7DAE5ULLATSNJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/G67C5S42GUMHL7DAE5ULLATSNJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/G67C5S42GUMHL7DAE5ULLATSNJ/action/storage_attestation","attest_author":"https://pith.science/pith/G67C5S42GUMHL7DAE5ULLATSNJ/action/author_attestation","sign_citation":"https://pith.science/pith/G67C5S42GUMHL7DAE5ULLATSNJ/action/citation_signature","submit_replication":"https://pith.science/pith/G67C5S42GUMHL7DAE5ULLATSNJ/action/replication_record"}},"created_at":"2026-08-04T02:06:09.865282+00:00","updated_at":"2026-08-04T02:06:09.865282+00:00"}