{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:U2OPI4VYJIFWGLUZ2F253OV4TQ","short_pith_number":"pith:U2OPI4VY","canonical_record":{"source":{"id":"2212.09686","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-12-19T18:14:36Z","cross_cats_sorted":[],"title_canon_sha256":"a2869996ff8f562cfe849d45579252c1253d80261fe202c95082f3651407be61","abstract_canon_sha256":"3092b06f1670526ae2c7f432be6ddb7fbd9aa33aeefe22b803f0fdd33fbad176"},"schema_version":"1.0"},"canonical_sha256":"a69cf472b84a0b632e99d175ddbabc9c10f11895708937a3fd14f24e26b5fc0e","source":{"kind":"arxiv","id":"2212.09686","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2212.09686","created_at":"2026-07-05T06:23:57Z"},{"alias_kind":"arxiv_version","alias_value":"2212.09686v2","created_at":"2026-07-05T06:23:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2212.09686","created_at":"2026-07-05T06:23:57Z"},{"alias_kind":"pith_short_12","alias_value":"U2OPI4VYJIFW","created_at":"2026-07-05T06:23:57Z"},{"alias_kind":"pith_short_16","alias_value":"U2OPI4VYJIFWGLUZ","created_at":"2026-07-05T06:23:57Z"},{"alias_kind":"pith_short_8","alias_value":"U2OPI4VY","created_at":"2026-07-05T06:23:57Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:U2OPI4VYJIFWGLUZ2F253OV4TQ","target":"record","payload":{"canonical_record":{"source":{"id":"2212.09686","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-12-19T18:14:36Z","cross_cats_sorted":[],"title_canon_sha256":"a2869996ff8f562cfe849d45579252c1253d80261fe202c95082f3651407be61","abstract_canon_sha256":"3092b06f1670526ae2c7f432be6ddb7fbd9aa33aeefe22b803f0fdd33fbad176"},"schema_version":"1.0"},"canonical_sha256":"a69cf472b84a0b632e99d175ddbabc9c10f11895708937a3fd14f24e26b5fc0e","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:23:57.514036Z","signature_b64":"HyzeppfD/74VZU19UT3JBG+YZrqSKdzBfaFVCa6LwAgmsvJpNCPStmOmHCzKyRTat+IsxwY2bY6Rfr5JLUq8Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a69cf472b84a0b632e99d175ddbabc9c10f11895708937a3fd14f24e26b5fc0e","last_reissued_at":"2026-07-05T06:23:57.513489Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:23:57.513489Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2212.09686","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":"2026-07-05T06:23:57Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"uRbp5TzT6paCcAlmh3Kq0orr++YVnMKcDNgkM1zuEdrcY3xXN3VffzjLWL/wxFe0E7r4AvdWsqztIbfiVxhqDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T10:40:55.049616Z"},"content_sha256":"7713fac7bc15044521c245e823a377ab7edd45d85d3fe10f69b0f8a0ca01bf6b","schema_version":"1.0","event_id":"sha256:7713fac7bc15044521c245e823a377ab7edd45d85d3fe10f69b0f8a0ca01bf6b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:U2OPI4VYJIFWGLUZ2F253OV4TQ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"A Natural Bias for Language Generation Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Adhiguna Kuncoro, Clara Meister, Laura Rimell, Lei Yu, Tiago Pimentel, Wojciech Stokowiec","submitted_at":"2022-12-19T18:14:36Z","abstract_excerpt":"After just a few hundred training updates, a standard probabilistic model for language generation has likely not yet learnt many semantic or syntactic rules of natural language, making it difficult to estimate the probability distribution over next tokens. Yet around this point, these models have identified a simple, loss-minimising behaviour: to output the unigram distribution of the target training corpus. The use of such a heuristic raises the question: Can we initialise our models with this behaviour and save precious compute resources and model capacity? Here we show that we can effective"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2212.09686","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/2212.09686/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":"2026-07-05T06:23:57Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gzEIyolCI6RIyRDwdGyx+5/8bjDCpWFLWs6R3E7aU657g4X2/k/JuryyJUs/+t4ydYhvTrxjefYsCruL4RN+Dw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T10:40:55.050162Z"},"content_sha256":"9f7e7c132fc0f5af4631dee5df12219c9ef6aa91e5809837c6e2afe744c45d0b","schema_version":"1.0","event_id":"sha256:9f7e7c132fc0f5af4631dee5df12219c9ef6aa91e5809837c6e2afe744c45d0b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/U2OPI4VYJIFWGLUZ2F253OV4TQ/bundle.json","state_url":"https://pith.science/pith/U2OPI4VYJIFWGLUZ2F253OV4TQ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/U2OPI4VYJIFWGLUZ2F253OV4TQ/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":"2026-08-13T10:40:55Z","links":{"resolver":"https://pith.science/pith/U2OPI4VYJIFWGLUZ2F253OV4TQ","bundle":"https://pith.science/pith/U2OPI4VYJIFWGLUZ2F253OV4TQ/bundle.json","state":"https://pith.science/pith/U2OPI4VYJIFWGLUZ2F253OV4TQ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/U2OPI4VYJIFWGLUZ2F253OV4TQ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:U2OPI4VYJIFWGLUZ2F253OV4TQ","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":"3092b06f1670526ae2c7f432be6ddb7fbd9aa33aeefe22b803f0fdd33fbad176","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-12-19T18:14:36Z","title_canon_sha256":"a2869996ff8f562cfe849d45579252c1253d80261fe202c95082f3651407be61"},"schema_version":"1.0","source":{"id":"2212.09686","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2212.09686","created_at":"2026-07-05T06:23:57Z"},{"alias_kind":"arxiv_version","alias_value":"2212.09686v2","created_at":"2026-07-05T06:23:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2212.09686","created_at":"2026-07-05T06:23:57Z"},{"alias_kind":"pith_short_12","alias_value":"U2OPI4VYJIFW","created_at":"2026-07-05T06:23:57Z"},{"alias_kind":"pith_short_16","alias_value":"U2OPI4VYJIFWGLUZ","created_at":"2026-07-05T06:23:57Z"},{"alias_kind":"pith_short_8","alias_value":"U2OPI4VY","created_at":"2026-07-05T06:23:57Z"}],"graph_snapshots":[{"event_id":"sha256:9f7e7c132fc0f5af4631dee5df12219c9ef6aa91e5809837c6e2afe744c45d0b","target":"graph","created_at":"2026-07-05T06:23:57Z","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/2212.09686/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"After just a few hundred training updates, a standard probabilistic model for language generation has likely not yet learnt many semantic or syntactic rules of natural language, making it difficult to estimate the probability distribution over next tokens. Yet around this point, these models have identified a simple, loss-minimising behaviour: to output the unigram distribution of the target training corpus. The use of such a heuristic raises the question: Can we initialise our models with this behaviour and save precious compute resources and model capacity? Here we show that we can effective","authors_text":"Adhiguna Kuncoro, Clara Meister, Laura Rimell, Lei Yu, Tiago Pimentel, Wojciech Stokowiec","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-12-19T18:14:36Z","title":"A Natural Bias for Language Generation Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2212.09686","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":"sha256:7713fac7bc15044521c245e823a377ab7edd45d85d3fe10f69b0f8a0ca01bf6b","target":"record","created_at":"2026-07-05T06:23:57Z","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":"3092b06f1670526ae2c7f432be6ddb7fbd9aa33aeefe22b803f0fdd33fbad176","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-12-19T18:14:36Z","title_canon_sha256":"a2869996ff8f562cfe849d45579252c1253d80261fe202c95082f3651407be61"},"schema_version":"1.0","source":{"id":"2212.09686","kind":"arxiv","version":2}},"canonical_sha256":"a69cf472b84a0b632e99d175ddbabc9c10f11895708937a3fd14f24e26b5fc0e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a69cf472b84a0b632e99d175ddbabc9c10f11895708937a3fd14f24e26b5fc0e","first_computed_at":"2026-07-05T06:23:57.513489Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:23:57.513489Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"HyzeppfD/74VZU19UT3JBG+YZrqSKdzBfaFVCa6LwAgmsvJpNCPStmOmHCzKyRTat+IsxwY2bY6Rfr5JLUq8Bg==","signature_status":"signed_v1","signed_at":"2026-07-05T06:23:57.514036Z","signed_message":"canonical_sha256_bytes"},"source_id":"2212.09686","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7713fac7bc15044521c245e823a377ab7edd45d85d3fe10f69b0f8a0ca01bf6b","sha256:9f7e7c132fc0f5af4631dee5df12219c9ef6aa91e5809837c6e2afe744c45d0b"],"state_sha256":"f1974e8d118cab6f2eaa3be01a96d77f6a9204ea7f652a8e6660299d16de1299"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Dh89QXQ1d9Zpvi3JIIZ+tP1vnAU07Keata8MG0oo2qnVruC/X7ayqFc9cgy0WMLaGFax0NOYsCqDrHjz4/NfDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-13T10:40:55.057496Z","bundle_sha256":"09f486e95a05e78082778cf8753899f931bd50095b649d4d1be17c660ffd0d2c"}}