{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:PX63XKQTP4IMHWPZBFOIQNBAWT","short_pith_number":"pith:PX63XKQT","canonical_record":{"source":{"id":"2301.02828","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-01-07T11:12:36Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"c862f30af65072b636e79c403fc5c7eba77e811273322d513ca21f87bc340c2e","abstract_canon_sha256":"7ac2096bc0d0c3ae518c2fd501ced877ececcb467c55021aef376e6e9704d7c7"},"schema_version":"1.0"},"canonical_sha256":"7dfdbbaa137f10c3d9f9095c883420b4f20350a565ad7f7b1f7e0a3c077fb997","source":{"kind":"arxiv","id":"2301.02828","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2301.02828","created_at":"2026-07-05T05:33:23Z"},{"alias_kind":"arxiv_version","alias_value":"2301.02828v2","created_at":"2026-07-05T05:33:23Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2301.02828","created_at":"2026-07-05T05:33:23Z"},{"alias_kind":"pith_short_12","alias_value":"PX63XKQTP4IM","created_at":"2026-07-05T05:33:23Z"},{"alias_kind":"pith_short_16","alias_value":"PX63XKQTP4IMHWPZ","created_at":"2026-07-05T05:33:23Z"},{"alias_kind":"pith_short_8","alias_value":"PX63XKQT","created_at":"2026-07-05T05:33:23Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:PX63XKQTP4IMHWPZBFOIQNBAWT","target":"record","payload":{"canonical_record":{"source":{"id":"2301.02828","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-01-07T11:12:36Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"c862f30af65072b636e79c403fc5c7eba77e811273322d513ca21f87bc340c2e","abstract_canon_sha256":"7ac2096bc0d0c3ae518c2fd501ced877ececcb467c55021aef376e6e9704d7c7"},"schema_version":"1.0"},"canonical_sha256":"7dfdbbaa137f10c3d9f9095c883420b4f20350a565ad7f7b1f7e0a3c077fb997","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:33:23.482672Z","signature_b64":"RYPcRmlP+vQWP+h+OJrQp9YWeUGBBSELP6U/Uj/CytIUN+woPjOzKgsCidCxxKDW5n9oUNiLBE32vgioC7VjDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7dfdbbaa137f10c3d9f9095c883420b4f20350a565ad7f7b1f7e0a3c077fb997","last_reissued_at":"2026-07-05T05:33:23.482258Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:33:23.482258Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2301.02828","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T05:33:23Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/nYd334MTYL8+dEWfJDCmqCtu/4EHpbXlFKc3CxgXrudaQTYM+0ND7aXBxQOm97tcbFZWlvJrgRObOxF9fOwDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T04:06:13.725457Z"},"content_sha256":"e8972a1c867dc6d8e7464bc9e88b887e9ba1530e5e77477b6492d6d1d28b408e","schema_version":"1.0","event_id":"sha256:e8972a1c867dc6d8e7464bc9e88b887e9ba1530e5e77477b6492d6d1d28b408e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:PX63XKQTP4IMHWPZBFOIQNBAWT","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Why do Nearest Neighbor Language Models Work?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Frank F. Xu, Graham Neubig, Uri Alon","submitted_at":"2023-01-07T11:12:36Z","abstract_excerpt":"Language models (LMs) compute the probability of a text by sequentially computing a representation of an already-seen context and using this representation to predict the next word. Currently, most LMs calculate these representations through a neural network consuming the immediate previous context. However recently, retrieval-augmented LMs have shown to improve over standard neural LMs, by accessing information retrieved from a large datastore, in addition to their standard, parametric, next-word prediction. In this paper, we set out to understand why retrieval-augmented language models, and "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2301.02828","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/2301.02828/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":"2026-07-05T05:33:23Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qMfeQIR9Abrsk3CK0rNdcASqBSMy1s3Omlnt/j218pD6yETeadEQvM+PbItaYltvzaRBTDRubI8JglQhDBmJAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T04:06:13.726385Z"},"content_sha256":"e4f9c1d131255cc79eeb0291dfb74518bd13a79ec9b1b5791c81bde3354822a0","schema_version":"1.0","event_id":"sha256:e4f9c1d131255cc79eeb0291dfb74518bd13a79ec9b1b5791c81bde3354822a0"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/PX63XKQTP4IMHWPZBFOIQNBAWT/bundle.json","state_url":"https://pith.science/pith/PX63XKQTP4IMHWPZBFOIQNBAWT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/PX63XKQTP4IMHWPZBFOIQNBAWT/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":"2026-08-13T04:06:13Z","links":{"resolver":"https://pith.science/pith/PX63XKQTP4IMHWPZBFOIQNBAWT","bundle":"https://pith.science/pith/PX63XKQTP4IMHWPZBFOIQNBAWT/bundle.json","state":"https://pith.science/pith/PX63XKQTP4IMHWPZBFOIQNBAWT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/PX63XKQTP4IMHWPZBFOIQNBAWT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:PX63XKQTP4IMHWPZBFOIQNBAWT","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":"7ac2096bc0d0c3ae518c2fd501ced877ececcb467c55021aef376e6e9704d7c7","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-01-07T11:12:36Z","title_canon_sha256":"c862f30af65072b636e79c403fc5c7eba77e811273322d513ca21f87bc340c2e"},"schema_version":"1.0","source":{"id":"2301.02828","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2301.02828","created_at":"2026-07-05T05:33:23Z"},{"alias_kind":"arxiv_version","alias_value":"2301.02828v2","created_at":"2026-07-05T05:33:23Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2301.02828","created_at":"2026-07-05T05:33:23Z"},{"alias_kind":"pith_short_12","alias_value":"PX63XKQTP4IM","created_at":"2026-07-05T05:33:23Z"},{"alias_kind":"pith_short_16","alias_value":"PX63XKQTP4IMHWPZ","created_at":"2026-07-05T05:33:23Z"},{"alias_kind":"pith_short_8","alias_value":"PX63XKQT","created_at":"2026-07-05T05:33:23Z"}],"graph_snapshots":[{"event_id":"sha256:e4f9c1d131255cc79eeb0291dfb74518bd13a79ec9b1b5791c81bde3354822a0","target":"graph","created_at":"2026-07-05T05:33:23Z","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/2301.02828/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Language models (LMs) compute the probability of a text by sequentially computing a representation of an already-seen context and using this representation to predict the next word. Currently, most LMs calculate these representations through a neural network consuming the immediate previous context. However recently, retrieval-augmented LMs have shown to improve over standard neural LMs, by accessing information retrieved from a large datastore, in addition to their standard, parametric, next-word prediction. In this paper, we set out to understand why retrieval-augmented language models, and ","authors_text":"Frank F. Xu, Graham Neubig, Uri Alon","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-01-07T11:12:36Z","title":"Why do Nearest Neighbor Language Models Work?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2301.02828","kind":"arxiv","version":2},"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:e8972a1c867dc6d8e7464bc9e88b887e9ba1530e5e77477b6492d6d1d28b408e","target":"record","created_at":"2026-07-05T05:33:23Z","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":"7ac2096bc0d0c3ae518c2fd501ced877ececcb467c55021aef376e6e9704d7c7","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-01-07T11:12:36Z","title_canon_sha256":"c862f30af65072b636e79c403fc5c7eba77e811273322d513ca21f87bc340c2e"},"schema_version":"1.0","source":{"id":"2301.02828","kind":"arxiv","version":2}},"canonical_sha256":"7dfdbbaa137f10c3d9f9095c883420b4f20350a565ad7f7b1f7e0a3c077fb997","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7dfdbbaa137f10c3d9f9095c883420b4f20350a565ad7f7b1f7e0a3c077fb997","first_computed_at":"2026-07-05T05:33:23.482258Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:33:23.482258Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"RYPcRmlP+vQWP+h+OJrQp9YWeUGBBSELP6U/Uj/CytIUN+woPjOzKgsCidCxxKDW5n9oUNiLBE32vgioC7VjDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T05:33:23.482672Z","signed_message":"canonical_sha256_bytes"},"source_id":"2301.02828","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e8972a1c867dc6d8e7464bc9e88b887e9ba1530e5e77477b6492d6d1d28b408e","sha256:e4f9c1d131255cc79eeb0291dfb74518bd13a79ec9b1b5791c81bde3354822a0"],"state_sha256":"03b609b88510636906bfeb2e6f0b16dafa2047dc01ba54b6df0f608cb2888179"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"p1V+14v7y8b5B58t5nfOBSq+ee6Nrz+jUU1Xyzk2lbgcT61jDGFOIsEr5Hz4X76RsuEvPSnl8rMn+dcvs6DdDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-13T04:06:13.733841Z","bundle_sha256":"ae83bf7d1468e1a6d2ca5242558b341f70805581a10c2e5aa425a4bd745bdef1"}}