{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:OKPEDTBAJWCIX2SD5YHDS3N3DY","short_pith_number":"pith:OKPEDTBA","canonical_record":{"source":{"id":"2310.14777","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-10-23T10:26:14Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"4ea84974271b5e8e192ac39f458a55ea6d3a72f68acc6461c2de26f9c7cc696a","abstract_canon_sha256":"b3c4a56d65547fc79e4c609723dbe70de276ed483e99031db743308e90c63340"},"schema_version":"1.0"},"canonical_sha256":"729e41cc204d848bea43ee0e396dbb1e09862ad2bee672db61b0d78585b2109a","source":{"kind":"arxiv","id":"2310.14777","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2310.14777","created_at":"2026-07-05T07:03:49Z"},{"alias_kind":"arxiv_version","alias_value":"2310.14777v1","created_at":"2026-07-05T07:03:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.14777","created_at":"2026-07-05T07:03:49Z"},{"alias_kind":"pith_short_12","alias_value":"OKPEDTBAJWCI","created_at":"2026-07-05T07:03:49Z"},{"alias_kind":"pith_short_16","alias_value":"OKPEDTBAJWCIX2SD","created_at":"2026-07-05T07:03:49Z"},{"alias_kind":"pith_short_8","alias_value":"OKPEDTBA","created_at":"2026-07-05T07:03:49Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:OKPEDTBAJWCIX2SD5YHDS3N3DY","target":"record","payload":{"canonical_record":{"source":{"id":"2310.14777","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-10-23T10:26:14Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"4ea84974271b5e8e192ac39f458a55ea6d3a72f68acc6461c2de26f9c7cc696a","abstract_canon_sha256":"b3c4a56d65547fc79e4c609723dbe70de276ed483e99031db743308e90c63340"},"schema_version":"1.0"},"canonical_sha256":"729e41cc204d848bea43ee0e396dbb1e09862ad2bee672db61b0d78585b2109a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:03:49.839698Z","signature_b64":"w3sAbBM3nZQ0uYpqJ6t18i6FQwntERjFuOs1yU0+8E/slpEfa9kq7OsFv2rcwr1VfINjSxzlhtho3+HEHtRKBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"729e41cc204d848bea43ee0e396dbb1e09862ad2bee672db61b0d78585b2109a","last_reissued_at":"2026-07-05T07:03:49.839266Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:03:49.839266Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2310.14777","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:03:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xOpkmd6zr+h6TkMy7S/W2sTDfeNFlnPyu1xPdZV3vlhbgDd2uKrW897tZ5R+h0vbx0IpQa2c0wKtOdI5+D8EAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T08:27:16.810533Z"},"content_sha256":"8b99f5714b70ce21ad1e0ffb0207109bee269551df2c4c7fe171e04f93177634","schema_version":"1.0","event_id":"sha256:8b99f5714b70ce21ad1e0ffb0207109bee269551df2c4c7fe171e04f93177634"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:OKPEDTBAJWCIX2SD5YHDS3N3DY","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Geographical Erasure in Language Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"C\\'edric Archambeau, Danish Pruthi, Jacek Golebiowski, Michele Donini, Pola Schw\\\"obel","submitted_at":"2023-10-23T10:26:14Z","abstract_excerpt":"Large language models (LLMs) encode vast amounts of world knowledge. However, since these models are trained on large swaths of internet data, they are at risk of inordinately capturing information about dominant groups. This imbalance can propagate into generated language. In this work, we study and operationalise a form of geographical erasure, wherein language models underpredict certain countries. We demonstrate consistent instances of erasure across a range of LLMs. We discover that erasure strongly correlates with low frequencies of country mentions in the training corpus. Lastly, we mit"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.14777","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/2310.14777/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":"2026-07-05T07:03:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"4rNDy6t/cI7iZcLeDNGBH7Vr0Aa8oFJggQd2/IpCJOiQsRTUpd70Ex7Hz3qJfIWUfukTz1V+ze+dHc98le49AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T08:27:16.810901Z"},"content_sha256":"42f9145bc30d015f7e6f65e8d37ad14792c50c622c2387bf43292bbee2b979b6","schema_version":"1.0","event_id":"sha256:42f9145bc30d015f7e6f65e8d37ad14792c50c622c2387bf43292bbee2b979b6"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/OKPEDTBAJWCIX2SD5YHDS3N3DY/bundle.json","state_url":"https://pith.science/pith/OKPEDTBAJWCIX2SD5YHDS3N3DY/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/OKPEDTBAJWCIX2SD5YHDS3N3DY/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":"2026-07-07T08:27:16Z","links":{"resolver":"https://pith.science/pith/OKPEDTBAJWCIX2SD5YHDS3N3DY","bundle":"https://pith.science/pith/OKPEDTBAJWCIX2SD5YHDS3N3DY/bundle.json","state":"https://pith.science/pith/OKPEDTBAJWCIX2SD5YHDS3N3DY/state.json","well_known_bundle":"https://pith.science/.well-known/pith/OKPEDTBAJWCIX2SD5YHDS3N3DY/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:OKPEDTBAJWCIX2SD5YHDS3N3DY","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":"b3c4a56d65547fc79e4c609723dbe70de276ed483e99031db743308e90c63340","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-10-23T10:26:14Z","title_canon_sha256":"4ea84974271b5e8e192ac39f458a55ea6d3a72f68acc6461c2de26f9c7cc696a"},"schema_version":"1.0","source":{"id":"2310.14777","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2310.14777","created_at":"2026-07-05T07:03:49Z"},{"alias_kind":"arxiv_version","alias_value":"2310.14777v1","created_at":"2026-07-05T07:03:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.14777","created_at":"2026-07-05T07:03:49Z"},{"alias_kind":"pith_short_12","alias_value":"OKPEDTBAJWCI","created_at":"2026-07-05T07:03:49Z"},{"alias_kind":"pith_short_16","alias_value":"OKPEDTBAJWCIX2SD","created_at":"2026-07-05T07:03:49Z"},{"alias_kind":"pith_short_8","alias_value":"OKPEDTBA","created_at":"2026-07-05T07:03:49Z"}],"graph_snapshots":[{"event_id":"sha256:42f9145bc30d015f7e6f65e8d37ad14792c50c622c2387bf43292bbee2b979b6","target":"graph","created_at":"2026-07-05T07:03:49Z","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/2310.14777/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) encode vast amounts of world knowledge. However, since these models are trained on large swaths of internet data, they are at risk of inordinately capturing information about dominant groups. This imbalance can propagate into generated language. In this work, we study and operationalise a form of geographical erasure, wherein language models underpredict certain countries. We demonstrate consistent instances of erasure across a range of LLMs. We discover that erasure strongly correlates with low frequencies of country mentions in the training corpus. Lastly, we mit","authors_text":"C\\'edric Archambeau, Danish Pruthi, Jacek Golebiowski, Michele Donini, Pola Schw\\\"obel","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-10-23T10:26:14Z","title":"Geographical Erasure in Language Generation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.14777","kind":"arxiv","version":1},"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":"sha256:8b99f5714b70ce21ad1e0ffb0207109bee269551df2c4c7fe171e04f93177634","target":"record","created_at":"2026-07-05T07:03:49Z","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":"b3c4a56d65547fc79e4c609723dbe70de276ed483e99031db743308e90c63340","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-10-23T10:26:14Z","title_canon_sha256":"4ea84974271b5e8e192ac39f458a55ea6d3a72f68acc6461c2de26f9c7cc696a"},"schema_version":"1.0","source":{"id":"2310.14777","kind":"arxiv","version":1}},"canonical_sha256":"729e41cc204d848bea43ee0e396dbb1e09862ad2bee672db61b0d78585b2109a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"729e41cc204d848bea43ee0e396dbb1e09862ad2bee672db61b0d78585b2109a","first_computed_at":"2026-07-05T07:03:49.839266Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:03:49.839266Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"w3sAbBM3nZQ0uYpqJ6t18i6FQwntERjFuOs1yU0+8E/slpEfa9kq7OsFv2rcwr1VfINjSxzlhtho3+HEHtRKBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T07:03:49.839698Z","signed_message":"canonical_sha256_bytes"},"source_id":"2310.14777","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8b99f5714b70ce21ad1e0ffb0207109bee269551df2c4c7fe171e04f93177634","sha256:42f9145bc30d015f7e6f65e8d37ad14792c50c622c2387bf43292bbee2b979b6"],"state_sha256":"4e7a4ca779a73854e545861e2ce696bc9b22ec69fa8c6464552d45f1e74da3b8"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SZiUM9vdjrgTZRQ1h5xJj3zDJuGLopkgqSk/WsrNUxXl8/eLCVquC0DTXh0WhLKVu1TjaYE1OOftig/IUeOGCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T08:27:16.812795Z","bundle_sha256":"9578246deb439f44968e6b8f9f9f0f68588741effa92e093e63abbde97e9cba2"}}