{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:HAHC26NZNKRMJ4HI7OOAGH2U7K","short_pith_number":"pith:HAHC26NZ","canonical_record":{"source":{"id":"2506.03100","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-03T17:31:53Z","cross_cats_sorted":["cs.AI","cs.CL","cs.IR","math.ST","stat.TH"],"title_canon_sha256":"87f4e9b86d30e44e7cb40aaf235873cd3aed049d576ae11420203dcf3685a55c","abstract_canon_sha256":"04510c21aa96665227a65029e9ac15767d4ef8fd64041118b1852ef9dccf71af"},"schema_version":"1.0"},"canonical_sha256":"380e2d79b96aa2c4f0e8fb9c031f54fa90e0dbcfa64f193b81063df072cfdb17","source":{"kind":"arxiv","id":"2506.03100","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.03100","created_at":"2026-07-05T11:18:22Z"},{"alias_kind":"arxiv_version","alias_value":"2506.03100v3","created_at":"2026-07-05T11:18:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.03100","created_at":"2026-07-05T11:18:22Z"},{"alias_kind":"pith_short_12","alias_value":"HAHC26NZNKRM","created_at":"2026-07-05T11:18:22Z"},{"alias_kind":"pith_short_16","alias_value":"HAHC26NZNKRMJ4HI","created_at":"2026-07-05T11:18:22Z"},{"alias_kind":"pith_short_8","alias_value":"HAHC26NZ","created_at":"2026-07-05T11:18:22Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:HAHC26NZNKRMJ4HI7OOAGH2U7K","target":"record","payload":{"canonical_record":{"source":{"id":"2506.03100","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-03T17:31:53Z","cross_cats_sorted":["cs.AI","cs.CL","cs.IR","math.ST","stat.TH"],"title_canon_sha256":"87f4e9b86d30e44e7cb40aaf235873cd3aed049d576ae11420203dcf3685a55c","abstract_canon_sha256":"04510c21aa96665227a65029e9ac15767d4ef8fd64041118b1852ef9dccf71af"},"schema_version":"1.0"},"canonical_sha256":"380e2d79b96aa2c4f0e8fb9c031f54fa90e0dbcfa64f193b81063df072cfdb17","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:18:22.852550Z","signature_b64":"m7rf/3LKPwmoELkrA25yALZ9i0fQWteALT2MeId/+ziIHcdGJDZVaqAWSeY4Mx/FQeUoQd5mCfz7deCmDLKdDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"380e2d79b96aa2c4f0e8fb9c031f54fa90e0dbcfa64f193b81063df072cfdb17","last_reissued_at":"2026-07-05T11:18:22.852067Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:18:22.852067Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.03100","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:18:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IrMBOzxrcssVNbZahCpCLAeBYgY2BqYx4iJnPuP0jZExYyUoUXG64eX0fHkVYfjuY4fpYLjkmm29xG7u/ZjtAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T07:45:49.128971Z"},"content_sha256":"d8adbb2f783f31252a170506cd0e3bbbc2110f9b06224286aee5e5f25c3bd19c","schema_version":"1.0","event_id":"sha256:d8adbb2f783f31252a170506cd0e3bbbc2110f9b06224286aee5e5f25c3bd19c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:HAHC26NZNKRMJ4HI7OOAGH2U7K","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Retrieval-Augmented Generation as Noisy In-Context Learning: A Unified Theory and Risk Bounds","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.IR","math.ST","stat.TH"],"primary_cat":"cs.LG","authors_text":"Robert D. Nowak, Yang Guo, Yifei Ming, Yingyu Liang, Yutian Tao","submitted_at":"2025-06-03T17:31:53Z","abstract_excerpt":"Retrieval-augmented generation (RAG) has seen many empirical successes in recent years by aiding the LLM with external knowledge. However, its theoretical aspect has remained mostly unexplored. In this paper, we propose the first finite-sample generalization bound for RAG in in-context linear regression and derive an exact bias-variance tradeoff. Our framework views the retrieved texts as query-dependent noisy in-context examples and recovers the classical in-context learning (ICL) and standard RAG as the limit cases. Our analysis suggests that an intrinsic ceiling on generalization error exis"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.03100","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/2506.03100/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:18:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OO7dzNIHY5rMaSUz/sFHUkIaRSkKZZiS715s6iBjwDpcYIvLw68H49yBVzHCZZaVLo25BmyuPueHtW0oM39MDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T07:45:49.129503Z"},"content_sha256":"21c919077822c7d467a9d0fe9ba9401400d73d7cd909c53bcc3a576728d93281","schema_version":"1.0","event_id":"sha256:21c919077822c7d467a9d0fe9ba9401400d73d7cd909c53bcc3a576728d93281"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HAHC26NZNKRMJ4HI7OOAGH2U7K/bundle.json","state_url":"https://pith.science/pith/HAHC26NZNKRMJ4HI7OOAGH2U7K/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HAHC26NZNKRMJ4HI7OOAGH2U7K/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-11T07:45:49Z","links":{"resolver":"https://pith.science/pith/HAHC26NZNKRMJ4HI7OOAGH2U7K","bundle":"https://pith.science/pith/HAHC26NZNKRMJ4HI7OOAGH2U7K/bundle.json","state":"https://pith.science/pith/HAHC26NZNKRMJ4HI7OOAGH2U7K/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HAHC26NZNKRMJ4HI7OOAGH2U7K/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:HAHC26NZNKRMJ4HI7OOAGH2U7K","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"04510c21aa96665227a65029e9ac15767d4ef8fd64041118b1852ef9dccf71af","cross_cats_sorted":["cs.AI","cs.CL","cs.IR","math.ST","stat.TH"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-03T17:31:53Z","title_canon_sha256":"87f4e9b86d30e44e7cb40aaf235873cd3aed049d576ae11420203dcf3685a55c"},"schema_version":"1.0","source":{"id":"2506.03100","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.03100","created_at":"2026-07-05T11:18:22Z"},{"alias_kind":"arxiv_version","alias_value":"2506.03100v3","created_at":"2026-07-05T11:18:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.03100","created_at":"2026-07-05T11:18:22Z"},{"alias_kind":"pith_short_12","alias_value":"HAHC26NZNKRM","created_at":"2026-07-05T11:18:22Z"},{"alias_kind":"pith_short_16","alias_value":"HAHC26NZNKRMJ4HI","created_at":"2026-07-05T11:18:22Z"},{"alias_kind":"pith_short_8","alias_value":"HAHC26NZ","created_at":"2026-07-05T11:18:22Z"}],"graph_snapshots":[{"event_id":"sha256:21c919077822c7d467a9d0fe9ba9401400d73d7cd909c53bcc3a576728d93281","target":"graph","created_at":"2026-07-05T11:18:22Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2506.03100/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Retrieval-augmented generation (RAG) has seen many empirical successes in recent years by aiding the LLM with external knowledge. However, its theoretical aspect has remained mostly unexplored. In this paper, we propose the first finite-sample generalization bound for RAG in in-context linear regression and derive an exact bias-variance tradeoff. Our framework views the retrieved texts as query-dependent noisy in-context examples and recovers the classical in-context learning (ICL) and standard RAG as the limit cases. Our analysis suggests that an intrinsic ceiling on generalization error exis","authors_text":"Robert D. Nowak, Yang Guo, Yifei Ming, Yingyu Liang, Yutian Tao","cross_cats":["cs.AI","cs.CL","cs.IR","math.ST","stat.TH"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-03T17:31:53Z","title":"Retrieval-Augmented Generation as Noisy In-Context Learning: A Unified Theory and Risk Bounds"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.03100","kind":"arxiv","version":3},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:d8adbb2f783f31252a170506cd0e3bbbc2110f9b06224286aee5e5f25c3bd19c","target":"record","created_at":"2026-07-05T11:18:22Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"04510c21aa96665227a65029e9ac15767d4ef8fd64041118b1852ef9dccf71af","cross_cats_sorted":["cs.AI","cs.CL","cs.IR","math.ST","stat.TH"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-03T17:31:53Z","title_canon_sha256":"87f4e9b86d30e44e7cb40aaf235873cd3aed049d576ae11420203dcf3685a55c"},"schema_version":"1.0","source":{"id":"2506.03100","kind":"arxiv","version":3}},"canonical_sha256":"380e2d79b96aa2c4f0e8fb9c031f54fa90e0dbcfa64f193b81063df072cfdb17","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"380e2d79b96aa2c4f0e8fb9c031f54fa90e0dbcfa64f193b81063df072cfdb17","first_computed_at":"2026-07-05T11:18:22.852067Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:18:22.852067Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"m7rf/3LKPwmoELkrA25yALZ9i0fQWteALT2MeId/+ziIHcdGJDZVaqAWSeY4Mx/FQeUoQd5mCfz7deCmDLKdDw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:18:22.852550Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.03100","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d8adbb2f783f31252a170506cd0e3bbbc2110f9b06224286aee5e5f25c3bd19c","sha256:21c919077822c7d467a9d0fe9ba9401400d73d7cd909c53bcc3a576728d93281"],"state_sha256":"34e837915b928abf07d7a551a061b246a54875ea2a9045bf5e547c3cdd116d1d"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"f6NuZiA6JILzPLWPxvvszFQ5OEvD+Mzjt6EKkUXBxd7Sk/qMG8Jk9Y+y+Hwa3H5xJTjdpaZNzYPndq83bP27AA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-11T07:45:49.134548Z","bundle_sha256":"976e83d6ea487d386f23265f0b84a873e71eeae9995b5823d70494ccce790a6a"}}