{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2020:QY4DZDTSM6OMXEIYYB3VGS3UVO","short_pith_number":"pith:QY4DZDTS","canonical_record":{"source":{"id":"2004.10964","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-04-23T04:21:19Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"133bc908d3b0ee1a7b2a66710754704e5c34b899f4d8d0aef42f455efa52056a","abstract_canon_sha256":"b3b003bde521fe5ef2661cc59275c825232844a96f47fb3a431550b32521caa2"},"schema_version":"1.0"},"canonical_sha256":"86383c8e72679ccb9118c077534b74ab9feeb0b44fcf61a556acd54f938adeaf","source":{"kind":"arxiv","id":"2004.10964","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2004.10964","created_at":"2026-07-05T01:00:51Z"},{"alias_kind":"arxiv_version","alias_value":"2004.10964v3","created_at":"2026-07-05T01:00:51Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2004.10964","created_at":"2026-07-05T01:00:51Z"},{"alias_kind":"pith_short_12","alias_value":"QY4DZDTSM6OM","created_at":"2026-07-05T01:00:51Z"},{"alias_kind":"pith_short_16","alias_value":"QY4DZDTSM6OMXEIY","created_at":"2026-07-05T01:00:51Z"},{"alias_kind":"pith_short_8","alias_value":"QY4DZDTS","created_at":"2026-07-05T01:00:51Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2020:QY4DZDTSM6OMXEIYYB3VGS3UVO","target":"record","payload":{"canonical_record":{"source":{"id":"2004.10964","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-04-23T04:21:19Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"133bc908d3b0ee1a7b2a66710754704e5c34b899f4d8d0aef42f455efa52056a","abstract_canon_sha256":"b3b003bde521fe5ef2661cc59275c825232844a96f47fb3a431550b32521caa2"},"schema_version":"1.0"},"canonical_sha256":"86383c8e72679ccb9118c077534b74ab9feeb0b44fcf61a556acd54f938adeaf","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:00:51.014032Z","signature_b64":"Nu3KSWtLvxUndm2yjFXdBypbh1cUO+JaRv+GJkXOdAB+lQYURUEZ5iuxIJy/r37rR6Toj1ir8/uxV5U04Lr6BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"86383c8e72679ccb9118c077534b74ab9feeb0b44fcf61a556acd54f938adeaf","last_reissued_at":"2026-07-05T01:00:51.013634Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:00:51.013634Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2004.10964","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":"2026-07-05T01:00:51Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"uNI20lzLEAZHT1gJxgPpUpryRrQ5+Fs1XN6wlhrJ7jyHM8s2saJZYmq0eqEk/Bimj/RfjK06NoLKLpbEwzB8AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T12:41:08.310610Z"},"content_sha256":"aee0a7f98e27d858baa29ed1b289aaf97b80bef1853aed1ee10f07cc049e48ee","schema_version":"1.0","event_id":"sha256:aee0a7f98e27d858baa29ed1b289aaf97b80bef1853aed1ee10f07cc049e48ee"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2020:QY4DZDTSM6OMXEIYYB3VGS3UVO","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Don't Stop Pretraining: Adapt Language Models to Domains and Tasks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Ana Marasovi\\'c, Doug Downey, Iz Beltagy, Kyle Lo, Noah A. Smith, Suchin Gururangan, Swabha Swayamdipta","submitted_at":"2020-04-23T04:21:19Z","abstract_excerpt":"Language models pretrained on text from a wide variety of sources form the foundation of today's NLP. In light of the success of these broad-coverage models, we investigate whether it is still helpful to tailor a pretrained model to the domain of a target task. We present a study across four domains (biomedical and computer science publications, news, and reviews) and eight classification tasks, showing that a second phase of pretraining in-domain (domain-adaptive pretraining) leads to performance gains, under both high- and low-resource settings. Moreover, adapting to the task's unlabeled dat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2004.10964","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/2004.10964/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":"2026-07-05T01:00:51Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Nrvup+0/WR+I3baNiOU4CYfGOITiJA/X8q8HJjFnqnKhti/3TFBJYbCQu/YPdlRjj1JjwD9TtCyUilwL+9kWCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T12:41:08.310990Z"},"content_sha256":"d64a8a0ba7a3572249bc84de380a4e9654ca0ac27406a7df902d5689c9828a4a","schema_version":"1.0","event_id":"sha256:d64a8a0ba7a3572249bc84de380a4e9654ca0ac27406a7df902d5689c9828a4a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QY4DZDTSM6OMXEIYYB3VGS3UVO/bundle.json","state_url":"https://pith.science/pith/QY4DZDTSM6OMXEIYYB3VGS3UVO/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QY4DZDTSM6OMXEIYYB3VGS3UVO/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":"2026-07-07T12:41:08Z","links":{"resolver":"https://pith.science/pith/QY4DZDTSM6OMXEIYYB3VGS3UVO","bundle":"https://pith.science/pith/QY4DZDTSM6OMXEIYYB3VGS3UVO/bundle.json","state":"https://pith.science/pith/QY4DZDTSM6OMXEIYYB3VGS3UVO/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QY4DZDTSM6OMXEIYYB3VGS3UVO/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:QY4DZDTSM6OMXEIYYB3VGS3UVO","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":"b3b003bde521fe5ef2661cc59275c825232844a96f47fb3a431550b32521caa2","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-04-23T04:21:19Z","title_canon_sha256":"133bc908d3b0ee1a7b2a66710754704e5c34b899f4d8d0aef42f455efa52056a"},"schema_version":"1.0","source":{"id":"2004.10964","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2004.10964","created_at":"2026-07-05T01:00:51Z"},{"alias_kind":"arxiv_version","alias_value":"2004.10964v3","created_at":"2026-07-05T01:00:51Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2004.10964","created_at":"2026-07-05T01:00:51Z"},{"alias_kind":"pith_short_12","alias_value":"QY4DZDTSM6OM","created_at":"2026-07-05T01:00:51Z"},{"alias_kind":"pith_short_16","alias_value":"QY4DZDTSM6OMXEIY","created_at":"2026-07-05T01:00:51Z"},{"alias_kind":"pith_short_8","alias_value":"QY4DZDTS","created_at":"2026-07-05T01:00:51Z"}],"graph_snapshots":[{"event_id":"sha256:d64a8a0ba7a3572249bc84de380a4e9654ca0ac27406a7df902d5689c9828a4a","target":"graph","created_at":"2026-07-05T01:00:51Z","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/2004.10964/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Language models pretrained on text from a wide variety of sources form the foundation of today's NLP. In light of the success of these broad-coverage models, we investigate whether it is still helpful to tailor a pretrained model to the domain of a target task. We present a study across four domains (biomedical and computer science publications, news, and reviews) and eight classification tasks, showing that a second phase of pretraining in-domain (domain-adaptive pretraining) leads to performance gains, under both high- and low-resource settings. Moreover, adapting to the task's unlabeled dat","authors_text":"Ana Marasovi\\'c, Doug Downey, Iz Beltagy, Kyle Lo, Noah A. Smith, Suchin Gururangan, Swabha Swayamdipta","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-04-23T04:21:19Z","title":"Don't Stop Pretraining: Adapt Language Models to Domains and Tasks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2004.10964","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:aee0a7f98e27d858baa29ed1b289aaf97b80bef1853aed1ee10f07cc049e48ee","target":"record","created_at":"2026-07-05T01:00:51Z","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":"b3b003bde521fe5ef2661cc59275c825232844a96f47fb3a431550b32521caa2","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-04-23T04:21:19Z","title_canon_sha256":"133bc908d3b0ee1a7b2a66710754704e5c34b899f4d8d0aef42f455efa52056a"},"schema_version":"1.0","source":{"id":"2004.10964","kind":"arxiv","version":3}},"canonical_sha256":"86383c8e72679ccb9118c077534b74ab9feeb0b44fcf61a556acd54f938adeaf","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"86383c8e72679ccb9118c077534b74ab9feeb0b44fcf61a556acd54f938adeaf","first_computed_at":"2026-07-05T01:00:51.013634Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T01:00:51.013634Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Nu3KSWtLvxUndm2yjFXdBypbh1cUO+JaRv+GJkXOdAB+lQYURUEZ5iuxIJy/r37rR6Toj1ir8/uxV5U04Lr6BA==","signature_status":"signed_v1","signed_at":"2026-07-05T01:00:51.014032Z","signed_message":"canonical_sha256_bytes"},"source_id":"2004.10964","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:aee0a7f98e27d858baa29ed1b289aaf97b80bef1853aed1ee10f07cc049e48ee","sha256:d64a8a0ba7a3572249bc84de380a4e9654ca0ac27406a7df902d5689c9828a4a"],"state_sha256":"76798a77bd9d1cbc6bd00481adc2cee47cacdbf7172bf75fea5e48e518e68e4c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Rlxgv2gC8EdoRSJ3AIbl42Om8ygVzdtVTJAtJD5TJDV6mSbF2FU7p+kLRZdcQ4qUaGERyHWYXBcN+C3pWN6eBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T12:41:08.312958Z","bundle_sha256":"1a9036d3909dd07a45c27ca9f44d63fb1fd0f6b02a98c7a7b4e0300112c1df0d"}}