{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:5LF47DFV4LLX26VGAGHTBKCT3M","short_pith_number":"pith:5LF47DFV","schema_version":"1.0","canonical_sha256":"eacbcf8cb5e2d77d7aa6018f30a853db0b3ea3054304e90d799d4307c152265e","source":{"kind":"arxiv","id":"2406.15319","version":3},"attestation_state":"computed","paper":{"title":"LongRAG: Enhancing Retrieval-Augmented Generation with Long-context LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Wenhu Chen, Xueguang Ma, Ziyan Jiang","submitted_at":"2024-06-21T17:23:21Z","abstract_excerpt":"In traditional RAG framework, the basic retrieval units are normally short. The common retrievers like DPR normally work with 100-word Wikipedia paragraphs. Such a design forces the retriever to search over a large corpus to find the `needle' unit. In contrast, the readers only need to generate answers from the short retrieved units. The imbalanced `heavy' retriever and `light' reader design can lead to sub-optimal performance. The loss of contextual information in the short, chunked units may increase the likelihood of introducing hard negatives during the retrieval stage. Additionally, the r"},"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":"2406.15319","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-21T17:23:21Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"92d52d9d155cec60decb81ad1ae910fac9323db70aded2b9965ea5314f1526be","abstract_canon_sha256":"13a8f0306ed7c5bf5b3b9904157cb1e1a2697775683c3a32632beec96dc3e5d1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:01:35.559516Z","signature_b64":"O6quca0pxQk3kT1/rw0PS1Xq2Z07TGxvc9ANSgV31mTxr1jg8xrgPthJZe9Ii73+1FHvhFKR2yxfK11edIrvBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"eacbcf8cb5e2d77d7aa6018f30a853db0b3ea3054304e90d799d4307c152265e","last_reissued_at":"2026-07-05T09:01:35.558981Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:01:35.558981Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LongRAG: Enhancing Retrieval-Augmented Generation with Long-context LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Wenhu Chen, Xueguang Ma, Ziyan Jiang","submitted_at":"2024-06-21T17:23:21Z","abstract_excerpt":"In traditional RAG framework, the basic retrieval units are normally short. The common retrievers like DPR normally work with 100-word Wikipedia paragraphs. Such a design forces the retriever to search over a large corpus to find the `needle' unit. In contrast, the readers only need to generate answers from the short retrieved units. The imbalanced `heavy' retriever and `light' reader design can lead to sub-optimal performance. The loss of contextual information in the short, chunked units may increase the likelihood of introducing hard negatives during the retrieval stage. Additionally, the r"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.15319","kind":"arxiv","version":3},"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/2406.15319/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":"2406.15319","created_at":"2026-07-05T09:01:35.559049+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.15319v3","created_at":"2026-07-05T09:01:35.559049+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.15319","created_at":"2026-07-05T09:01:35.559049+00:00"},{"alias_kind":"pith_short_12","alias_value":"5LF47DFV4LLX","created_at":"2026-07-05T09:01:35.559049+00:00"},{"alias_kind":"pith_short_16","alias_value":"5LF47DFV4LLX26VG","created_at":"2026-07-05T09:01:35.559049+00:00"},{"alias_kind":"pith_short_8","alias_value":"5LF47DFV","created_at":"2026-07-05T09:01:35.559049+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":16,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.13550","citing_title":"Uncertainty-Aware Hybrid Retrieval for Long-Document RAG","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2607.02073","citing_title":"Evidence-State Rewards for Long-Context Reasoning","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14906","citing_title":"MemLens: Benchmarking Multimodal Long-Term Memory in Large Vision-Language Models","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2606.28349","citing_title":"HMARS: A Hierarchical Multi-Agent Memory System for Long-Context Reasoning","ref_index":83,"is_internal_anchor":false},{"citing_arxiv_id":"2502.13957","citing_title":"Supervising the search process produces reliable and generalizable information-seeking agents","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2504.02181","citing_title":"A Survey of Scaling in Large Language Model Reasoning","ref_index":71,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18767","citing_title":"DualView: Adaptive Local-Global Fusion for Multi-Hop Document Reranking","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2507.15586","citing_title":"Learning to Extract Rational Evidence via Reinforcement Learning for Retrieval-Augmented Generation","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2603.26815","citing_title":"Sustainable Hybrid Document-Routed Retrieval for Financial RAG: Resolving the Robustness-Precision Trade-off","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2604.01432","citing_title":"Are Finer Citations Always Better? Rethinking Granularity for Attributed Generation","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09942","citing_title":"HAGE: Harnessing Agentic Memory via RL-Driven Weighted Graph Evolution","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2605.03989","citing_title":"An Agent-Oriented Pluggable Experience-RAG Skill for Experience-Driven Retrieval Strategy Orchestration","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2604.22861","citing_title":"IntrAgent: An LLM Agent for Content-Grounded Information Retrieval through Literature Review","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2604.11419","citing_title":"Beyond RAG for Cyber Threat Intelligence: A Systematic Evaluation of Graph-Based and Agentic Retrieval","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2604.15583","citing_title":"SAGE: Selective Attention-Guided Extraction for Token-Efficient Document Indexing","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2605.03989","citing_title":"An Agent-Oriented Pluggable Experience-RAG Skill for Experience-Driven Retrieval Strategy Orchestration","ref_index":10,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5LF47DFV4LLX26VGAGHTBKCT3M","json":"https://pith.science/pith/5LF47DFV4LLX26VGAGHTBKCT3M.json","graph_json":"https://pith.science/api/pith-number/5LF47DFV4LLX26VGAGHTBKCT3M/graph.json","events_json":"https://pith.science/api/pith-number/5LF47DFV4LLX26VGAGHTBKCT3M/events.json","paper":"https://pith.science/paper/5LF47DFV"},"agent_actions":{"view_html":"https://pith.science/pith/5LF47DFV4LLX26VGAGHTBKCT3M","download_json":"https://pith.science/pith/5LF47DFV4LLX26VGAGHTBKCT3M.json","view_paper":"https://pith.science/paper/5LF47DFV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.15319&json=true","fetch_graph":"https://pith.science/api/pith-number/5LF47DFV4LLX26VGAGHTBKCT3M/graph.json","fetch_events":"https://pith.science/api/pith-number/5LF47DFV4LLX26VGAGHTBKCT3M/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5LF47DFV4LLX26VGAGHTBKCT3M/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5LF47DFV4LLX26VGAGHTBKCT3M/action/storage_attestation","attest_author":"https://pith.science/pith/5LF47DFV4LLX26VGAGHTBKCT3M/action/author_attestation","sign_citation":"https://pith.science/pith/5LF47DFV4LLX26VGAGHTBKCT3M/action/citation_signature","submit_replication":"https://pith.science/pith/5LF47DFV4LLX26VGAGHTBKCT3M/action/replication_record"}},"created_at":"2026-07-05T09:01:35.559049+00:00","updated_at":"2026-07-05T09:01:35.559049+00:00"}