{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:Q73HBJG2UUCQ6HNMK3A5IZPWVG","short_pith_number":"pith:Q73HBJG2","schema_version":"1.0","canonical_sha256":"87f670a4daa5050f1dac56c1d465f6a9932006eedb96e5e7806d9ae35b510c7e","source":{"kind":"arxiv","id":"2502.09977","version":2},"attestation_state":"computed","paper":{"title":"LaRA: Benchmarking Retrieval-Augmented Generation and Long-Context LLMs -- No Silver Bullet for LC or RAG Routing","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Fei Huang, Kuan Li, Liwen Zhang, Minhao Cheng, Pengjun Xie, Shuai Wang, Yong Jiang","submitted_at":"2025-02-14T08:04:22Z","abstract_excerpt":"Effectively incorporating external knowledge into Large Language Models (LLMs) is crucial for enhancing their capabilities and addressing real-world needs. Retrieval-Augmented Generation (RAG) offers an effective method for achieving this by retrieving the most relevant fragments into LLMs. However, the advancements in context window size for LLMs offer an alternative approach, raising the question of whether RAG remains necessary for effectively handling external knowledge. Several existing studies provide inconclusive comparisons between RAG and long-context (LC) LLMs, largely due to limitat"},"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":"2502.09977","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-14T08:04:22Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"fa2e3332efbf6c93c4d969dfa07d6b507318723654b65830515c34f52e47fd7d","abstract_canon_sha256":"788eb13fa951eb843a30a4ff2bb0af5ab4db89b1696799e20775a8bb096f4c2c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:24:31.179588Z","signature_b64":"3kkKcnuPxD/1+XJup0rAIYRdy+6r2w9cW32y5zAEKnaw5o4uLAD7kqan9eDQ137f4kLoPL865Fbq4aWeBSuoAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"87f670a4daa5050f1dac56c1d465f6a9932006eedb96e5e7806d9ae35b510c7e","last_reissued_at":"2026-07-05T10:24:31.178919Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:24:31.178919Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LaRA: Benchmarking Retrieval-Augmented Generation and Long-Context LLMs -- No Silver Bullet for LC or RAG Routing","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Fei Huang, Kuan Li, Liwen Zhang, Minhao Cheng, Pengjun Xie, Shuai Wang, Yong Jiang","submitted_at":"2025-02-14T08:04:22Z","abstract_excerpt":"Effectively incorporating external knowledge into Large Language Models (LLMs) is crucial for enhancing their capabilities and addressing real-world needs. Retrieval-Augmented Generation (RAG) offers an effective method for achieving this by retrieving the most relevant fragments into LLMs. However, the advancements in context window size for LLMs offer an alternative approach, raising the question of whether RAG remains necessary for effectively handling external knowledge. Several existing studies provide inconclusive comparisons between RAG and long-context (LC) LLMs, largely due to limitat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.09977","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/2502.09977/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":"2502.09977","created_at":"2026-07-05T10:24:31.178989+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.09977v2","created_at":"2026-07-05T10:24:31.178989+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.09977","created_at":"2026-07-05T10:24:31.178989+00:00"},{"alias_kind":"pith_short_12","alias_value":"Q73HBJG2UUCQ","created_at":"2026-07-05T10:24:31.178989+00:00"},{"alias_kind":"pith_short_16","alias_value":"Q73HBJG2UUCQ6HNM","created_at":"2026-07-05T10:24:31.178989+00:00"},{"alias_kind":"pith_short_8","alias_value":"Q73HBJG2","created_at":"2026-07-05T10:24:31.178989+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.10235","citing_title":"Route Before Retrieve: Activating Latent Routing Abilities of LLMs for RAG vs. Long-Context Selection","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10235","citing_title":"Route Before Retrieve: Activating Latent Routing Abilities of LLMs for RAG vs. Long-Context Selection","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02173","citing_title":"Retrieval and Multi-Hop Reasoning in 1M-Token Context Windows: Evaluating LLMs on Classical Chinese Text","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Q73HBJG2UUCQ6HNMK3A5IZPWVG","json":"https://pith.science/pith/Q73HBJG2UUCQ6HNMK3A5IZPWVG.json","graph_json":"https://pith.science/api/pith-number/Q73HBJG2UUCQ6HNMK3A5IZPWVG/graph.json","events_json":"https://pith.science/api/pith-number/Q73HBJG2UUCQ6HNMK3A5IZPWVG/events.json","paper":"https://pith.science/paper/Q73HBJG2"},"agent_actions":{"view_html":"https://pith.science/pith/Q73HBJG2UUCQ6HNMK3A5IZPWVG","download_json":"https://pith.science/pith/Q73HBJG2UUCQ6HNMK3A5IZPWVG.json","view_paper":"https://pith.science/paper/Q73HBJG2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.09977&json=true","fetch_graph":"https://pith.science/api/pith-number/Q73HBJG2UUCQ6HNMK3A5IZPWVG/graph.json","fetch_events":"https://pith.science/api/pith-number/Q73HBJG2UUCQ6HNMK3A5IZPWVG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Q73HBJG2UUCQ6HNMK3A5IZPWVG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Q73HBJG2UUCQ6HNMK3A5IZPWVG/action/storage_attestation","attest_author":"https://pith.science/pith/Q73HBJG2UUCQ6HNMK3A5IZPWVG/action/author_attestation","sign_citation":"https://pith.science/pith/Q73HBJG2UUCQ6HNMK3A5IZPWVG/action/citation_signature","submit_replication":"https://pith.science/pith/Q73HBJG2UUCQ6HNMK3A5IZPWVG/action/replication_record"}},"created_at":"2026-07-05T10:24:31.178989+00:00","updated_at":"2026-07-05T10:24:31.178989+00:00"}