{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:T5RTKUS3PH42JRC65ZVTI2JTIO","short_pith_number":"pith:T5RTKUS3","canonical_record":{"source":{"id":"2111.09509","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-11-18T04:07:09Z","cross_cats_sorted":[],"title_canon_sha256":"a1c629a351f0d7dbe738f6dde2eeaa5a6850a7d7283f7af81373cc9e2511dd81","abstract_canon_sha256":"bd0eafad71f48f5fef6a1c1b4a18c4fea7caf412bab2de10c94698fc27cab089"},"schema_version":"1.0"},"canonical_sha256":"9f6335525b79f9a4c45eee6b346933438b32f15c2663897784eba4ea010ff709","source":{"kind":"arxiv","id":"2111.09509","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2111.09509","created_at":"2026-07-05T03:33:12Z"},{"alias_kind":"arxiv_version","alias_value":"2111.09509v1","created_at":"2026-07-05T03:33:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2111.09509","created_at":"2026-07-05T03:33:12Z"},{"alias_kind":"pith_short_12","alias_value":"T5RTKUS3PH42","created_at":"2026-07-05T03:33:12Z"},{"alias_kind":"pith_short_16","alias_value":"T5RTKUS3PH42JRC6","created_at":"2026-07-05T03:33:12Z"},{"alias_kind":"pith_short_8","alias_value":"T5RTKUS3","created_at":"2026-07-05T03:33:12Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:T5RTKUS3PH42JRC65ZVTI2JTIO","target":"record","payload":{"canonical_record":{"source":{"id":"2111.09509","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-11-18T04:07:09Z","cross_cats_sorted":[],"title_canon_sha256":"a1c629a351f0d7dbe738f6dde2eeaa5a6850a7d7283f7af81373cc9e2511dd81","abstract_canon_sha256":"bd0eafad71f48f5fef6a1c1b4a18c4fea7caf412bab2de10c94698fc27cab089"},"schema_version":"1.0"},"canonical_sha256":"9f6335525b79f9a4c45eee6b346933438b32f15c2663897784eba4ea010ff709","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:33:12.409371Z","signature_b64":"43Zk0bYuagmu9mhFNu3TA+5uF6HY81/x3L0MYNPAXavmwUhUUUlGZWCcHalDldd37bcSR9xZ2UDI519GSzOOAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9f6335525b79f9a4c45eee6b346933438b32f15c2663897784eba4ea010ff709","last_reissued_at":"2026-07-05T03:33:12.408960Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:33:12.408960Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2111.09509","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":"2026-07-05T03:33:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Yazyw0kIbS3VeI1YJXr6bpZOdfGYSzBfa6w0I0Fc2pP+Sc+2b8mMm6l0cDvctCDUOJDiTBhF27UnYBxsudnwCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T08:22:30.699807Z"},"content_sha256":"cfbcc1bd84855f4fe1ebab01aa68060b37ccaab6cabb0be8b3b9382173823257","schema_version":"1.0","event_id":"sha256:cfbcc1bd84855f4fe1ebab01aa68060b37ccaab6cabb0be8b3b9382173823257"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:T5RTKUS3PH42JRC65ZVTI2JTIO","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"How much do language models copy from their training data? Evaluating linguistic novelty in text generation using RAVEN","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Asli Celikyilmaz, Jianfeng Gao, Paul Smolensky, R. Thomas McCoy, Tal Linzen","submitted_at":"2021-11-18T04:07:09Z","abstract_excerpt":"Current language models can generate high-quality text. Are they simply copying text they have seen before, or have they learned generalizable linguistic abstractions? To tease apart these possibilities, we introduce RAVEN, a suite of analyses for assessing the novelty of generated text, focusing on sequential structure (n-grams) and syntactic structure. We apply these analyses to four neural language models (an LSTM, a Transformer, Transformer-XL, and GPT-2). For local structure - e.g., individual dependencies - model-generated text is substantially less novel than our baseline of human-gener"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2111.09509","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/2111.09509/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":"2026-07-05T03:33:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"enjDZSUqySSG4LQetB/TgkhA0ozRDgdTvjK6AL1gnMJlLKRlnUNhCvb6kISHjVQBd+DobvxdFes2TgrhUSaLBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T08:22:30.700304Z"},"content_sha256":"ca574f6cdfc5abd8694546c78de7444f0e03a92e3669dbc6e0e8c196ef33538c","schema_version":"1.0","event_id":"sha256:ca574f6cdfc5abd8694546c78de7444f0e03a92e3669dbc6e0e8c196ef33538c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/T5RTKUS3PH42JRC65ZVTI2JTIO/bundle.json","state_url":"https://pith.science/pith/T5RTKUS3PH42JRC65ZVTI2JTIO/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/T5RTKUS3PH42JRC65ZVTI2JTIO/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":"2026-08-16T08:22:30Z","links":{"resolver":"https://pith.science/pith/T5RTKUS3PH42JRC65ZVTI2JTIO","bundle":"https://pith.science/pith/T5RTKUS3PH42JRC65ZVTI2JTIO/bundle.json","state":"https://pith.science/pith/T5RTKUS3PH42JRC65ZVTI2JTIO/state.json","well_known_bundle":"https://pith.science/.well-known/pith/T5RTKUS3PH42JRC65ZVTI2JTIO/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:T5RTKUS3PH42JRC65ZVTI2JTIO","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":"bd0eafad71f48f5fef6a1c1b4a18c4fea7caf412bab2de10c94698fc27cab089","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-11-18T04:07:09Z","title_canon_sha256":"a1c629a351f0d7dbe738f6dde2eeaa5a6850a7d7283f7af81373cc9e2511dd81"},"schema_version":"1.0","source":{"id":"2111.09509","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2111.09509","created_at":"2026-07-05T03:33:12Z"},{"alias_kind":"arxiv_version","alias_value":"2111.09509v1","created_at":"2026-07-05T03:33:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2111.09509","created_at":"2026-07-05T03:33:12Z"},{"alias_kind":"pith_short_12","alias_value":"T5RTKUS3PH42","created_at":"2026-07-05T03:33:12Z"},{"alias_kind":"pith_short_16","alias_value":"T5RTKUS3PH42JRC6","created_at":"2026-07-05T03:33:12Z"},{"alias_kind":"pith_short_8","alias_value":"T5RTKUS3","created_at":"2026-07-05T03:33:12Z"}],"graph_snapshots":[{"event_id":"sha256:ca574f6cdfc5abd8694546c78de7444f0e03a92e3669dbc6e0e8c196ef33538c","target":"graph","created_at":"2026-07-05T03:33:12Z","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/2111.09509/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Current language models can generate high-quality text. Are they simply copying text they have seen before, or have they learned generalizable linguistic abstractions? To tease apart these possibilities, we introduce RAVEN, a suite of analyses for assessing the novelty of generated text, focusing on sequential structure (n-grams) and syntactic structure. We apply these analyses to four neural language models (an LSTM, a Transformer, Transformer-XL, and GPT-2). For local structure - e.g., individual dependencies - model-generated text is substantially less novel than our baseline of human-gener","authors_text":"Asli Celikyilmaz, Jianfeng Gao, Paul Smolensky, R. Thomas McCoy, Tal Linzen","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-11-18T04:07:09Z","title":"How much do language models copy from their training data? Evaluating linguistic novelty in text generation using RAVEN"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2111.09509","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:cfbcc1bd84855f4fe1ebab01aa68060b37ccaab6cabb0be8b3b9382173823257","target":"record","created_at":"2026-07-05T03:33:12Z","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":"bd0eafad71f48f5fef6a1c1b4a18c4fea7caf412bab2de10c94698fc27cab089","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-11-18T04:07:09Z","title_canon_sha256":"a1c629a351f0d7dbe738f6dde2eeaa5a6850a7d7283f7af81373cc9e2511dd81"},"schema_version":"1.0","source":{"id":"2111.09509","kind":"arxiv","version":1}},"canonical_sha256":"9f6335525b79f9a4c45eee6b346933438b32f15c2663897784eba4ea010ff709","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9f6335525b79f9a4c45eee6b346933438b32f15c2663897784eba4ea010ff709","first_computed_at":"2026-07-05T03:33:12.408960Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T03:33:12.408960Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"43Zk0bYuagmu9mhFNu3TA+5uF6HY81/x3L0MYNPAXavmwUhUUUlGZWCcHalDldd37bcSR9xZ2UDI519GSzOOAg==","signature_status":"signed_v1","signed_at":"2026-07-05T03:33:12.409371Z","signed_message":"canonical_sha256_bytes"},"source_id":"2111.09509","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:cfbcc1bd84855f4fe1ebab01aa68060b37ccaab6cabb0be8b3b9382173823257","sha256:ca574f6cdfc5abd8694546c78de7444f0e03a92e3669dbc6e0e8c196ef33538c"],"state_sha256":"2d12bf921f1a0092119535e71eeee1e6d79997ac6284a2d13930e5f5b771362b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SUNi2Z9pnIkc4WFJZNR+y8n0Pr3iJNGDZ2er1GDXrW/QFzclTYV3niAmbaxesu/MyFlBRTTZSph0h4TbYTYLDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-16T08:22:30.704918Z","bundle_sha256":"c2aebc9a5e43e97cf312125b3e359277cf8ca53eb65106c57ec99db016616505"}}