{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2020:4TH77EWDUU4JT55XWE3FIF4KEI","short_pith_number":"pith:4TH77EWD","canonical_record":{"source":{"id":"2010.05904","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-10-12T17:57:00Z","cross_cats_sorted":[],"title_canon_sha256":"5d45da4876116e53b87fa634f9d7793211ce8e4269fe27ef76552260ba5959ea","abstract_canon_sha256":"64139c81c0de7fbf1989760bd5b214056c13c5df3e7b05cd68cb01cd51b8839a"},"schema_version":"1.0"},"canonical_sha256":"e4cfff92c3a53899f7b7b13654178a2228cc8e9366b1ee991ff99061940734f3","source":{"kind":"arxiv","id":"2010.05904","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2010.05904","created_at":"2026-07-05T01:42:11Z"},{"alias_kind":"arxiv_version","alias_value":"2010.05904v1","created_at":"2026-07-05T01:42:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2010.05904","created_at":"2026-07-05T01:42:11Z"},{"alias_kind":"pith_short_12","alias_value":"4TH77EWDUU4J","created_at":"2026-07-05T01:42:11Z"},{"alias_kind":"pith_short_16","alias_value":"4TH77EWDUU4JT55X","created_at":"2026-07-05T01:42:11Z"},{"alias_kind":"pith_short_8","alias_value":"4TH77EWD","created_at":"2026-07-05T01:42:11Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2020:4TH77EWDUU4JT55XWE3FIF4KEI","target":"record","payload":{"canonical_record":{"source":{"id":"2010.05904","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-10-12T17:57:00Z","cross_cats_sorted":[],"title_canon_sha256":"5d45da4876116e53b87fa634f9d7793211ce8e4269fe27ef76552260ba5959ea","abstract_canon_sha256":"64139c81c0de7fbf1989760bd5b214056c13c5df3e7b05cd68cb01cd51b8839a"},"schema_version":"1.0"},"canonical_sha256":"e4cfff92c3a53899f7b7b13654178a2228cc8e9366b1ee991ff99061940734f3","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:42:11.583558Z","signature_b64":"vIvtkDVVHLm3yMLGb0nnXZiaSBm4HAE/TMPdCJynLn1slkQ4Bd7Yc/e8MkME71pZ2xXF7QdOOB4aWgK4kTGJBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e4cfff92c3a53899f7b7b13654178a2228cc8e9366b1ee991ff99061940734f3","last_reissued_at":"2026-07-05T01:42:11.583117Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:42:11.583117Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2010.05904","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":"2026-07-05T01:42:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"iSJH3pZiHJBxMuNFz0ZZWfmZogHtMOJAwM5UgEdjlMYxO5ED3lojpVDnXTQ7CfdRqlOkcGar5Flk6ukmMghoCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T21:49:08.102484Z"},"content_sha256":"d34ef02478f581abcde2e53a65b6e235cfa5b87dba53608299a47145eb0775c6","schema_version":"1.0","event_id":"sha256:d34ef02478f581abcde2e53a65b6e235cfa5b87dba53608299a47145eb0775c6"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2020:4TH77EWDUU4JT55XWE3FIF4KEI","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Multi-Stage Pre-training for Low-Resource Domain Adaptation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Anthony Ferritto, Avirup Sil, Efsun Sarioglu Kayi, Md Arafat Sultan, Radu Florian, Revanth Gangi Reddy, Rong Zhang, Salim Roukos, Todd Ward, Vittorio Castelli","submitted_at":"2020-10-12T17:57:00Z","abstract_excerpt":"Transfer learning techniques are particularly useful in NLP tasks where a sizable amount of high-quality annotated data is difficult to obtain. Current approaches directly adapt a pre-trained language model (LM) on in-domain text before fine-tuning to downstream tasks. We show that extending the vocabulary of the LM with domain-specific terms leads to further gains. To a bigger effect, we utilize structure in the unlabeled data to create auxiliary synthetic tasks, which helps the LM transfer to downstream tasks. We apply these approaches incrementally on a pre-trained Roberta-large LM and show"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2010.05904","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/2010.05904/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":"2026-07-05T01:42:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"hPY8qynobGT6arMIbZFAJ2LDShjuKhlHvWq+3+re+3269EUVqB2xYa4De+VKU6SgQP4C9KqZcPHe6/JV4rWGBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T21:49:08.102855Z"},"content_sha256":"090f4fa890ab765cae829f5eae05871f45f0e9a2cd7bc1b7e1102581b435de24","schema_version":"1.0","event_id":"sha256:090f4fa890ab765cae829f5eae05871f45f0e9a2cd7bc1b7e1102581b435de24"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4TH77EWDUU4JT55XWE3FIF4KEI/bundle.json","state_url":"https://pith.science/pith/4TH77EWDUU4JT55XWE3FIF4KEI/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4TH77EWDUU4JT55XWE3FIF4KEI/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":"2026-07-06T21:49:08Z","links":{"resolver":"https://pith.science/pith/4TH77EWDUU4JT55XWE3FIF4KEI","bundle":"https://pith.science/pith/4TH77EWDUU4JT55XWE3FIF4KEI/bundle.json","state":"https://pith.science/pith/4TH77EWDUU4JT55XWE3FIF4KEI/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4TH77EWDUU4JT55XWE3FIF4KEI/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:4TH77EWDUU4JT55XWE3FIF4KEI","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":"64139c81c0de7fbf1989760bd5b214056c13c5df3e7b05cd68cb01cd51b8839a","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-10-12T17:57:00Z","title_canon_sha256":"5d45da4876116e53b87fa634f9d7793211ce8e4269fe27ef76552260ba5959ea"},"schema_version":"1.0","source":{"id":"2010.05904","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2010.05904","created_at":"2026-07-05T01:42:11Z"},{"alias_kind":"arxiv_version","alias_value":"2010.05904v1","created_at":"2026-07-05T01:42:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2010.05904","created_at":"2026-07-05T01:42:11Z"},{"alias_kind":"pith_short_12","alias_value":"4TH77EWDUU4J","created_at":"2026-07-05T01:42:11Z"},{"alias_kind":"pith_short_16","alias_value":"4TH77EWDUU4JT55X","created_at":"2026-07-05T01:42:11Z"},{"alias_kind":"pith_short_8","alias_value":"4TH77EWD","created_at":"2026-07-05T01:42:11Z"}],"graph_snapshots":[{"event_id":"sha256:090f4fa890ab765cae829f5eae05871f45f0e9a2cd7bc1b7e1102581b435de24","target":"graph","created_at":"2026-07-05T01:42:11Z","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/2010.05904/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Transfer learning techniques are particularly useful in NLP tasks where a sizable amount of high-quality annotated data is difficult to obtain. Current approaches directly adapt a pre-trained language model (LM) on in-domain text before fine-tuning to downstream tasks. We show that extending the vocabulary of the LM with domain-specific terms leads to further gains. To a bigger effect, we utilize structure in the unlabeled data to create auxiliary synthetic tasks, which helps the LM transfer to downstream tasks. We apply these approaches incrementally on a pre-trained Roberta-large LM and show","authors_text":"Anthony Ferritto, Avirup Sil, Efsun Sarioglu Kayi, Md Arafat Sultan, Radu Florian, Revanth Gangi Reddy, Rong Zhang, Salim Roukos, Todd Ward, Vittorio Castelli","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-10-12T17:57:00Z","title":"Multi-Stage Pre-training for Low-Resource Domain Adaptation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2010.05904","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:d34ef02478f581abcde2e53a65b6e235cfa5b87dba53608299a47145eb0775c6","target":"record","created_at":"2026-07-05T01:42:11Z","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":"64139c81c0de7fbf1989760bd5b214056c13c5df3e7b05cd68cb01cd51b8839a","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-10-12T17:57:00Z","title_canon_sha256":"5d45da4876116e53b87fa634f9d7793211ce8e4269fe27ef76552260ba5959ea"},"schema_version":"1.0","source":{"id":"2010.05904","kind":"arxiv","version":1}},"canonical_sha256":"e4cfff92c3a53899f7b7b13654178a2228cc8e9366b1ee991ff99061940734f3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e4cfff92c3a53899f7b7b13654178a2228cc8e9366b1ee991ff99061940734f3","first_computed_at":"2026-07-05T01:42:11.583117Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T01:42:11.583117Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"vIvtkDVVHLm3yMLGb0nnXZiaSBm4HAE/TMPdCJynLn1slkQ4Bd7Yc/e8MkME71pZ2xXF7QdOOB4aWgK4kTGJBA==","signature_status":"signed_v1","signed_at":"2026-07-05T01:42:11.583558Z","signed_message":"canonical_sha256_bytes"},"source_id":"2010.05904","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d34ef02478f581abcde2e53a65b6e235cfa5b87dba53608299a47145eb0775c6","sha256:090f4fa890ab765cae829f5eae05871f45f0e9a2cd7bc1b7e1102581b435de24"],"state_sha256":"30116821b4e24eb07ebbf7a34d281fa00e9377d6b61b962c24e25b5aa9c2d5fc"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Yq6BMBYlyvxBDhhIgyLqwGqTcje6u0GEIrnpRSnWTk49fBNyW1esZTZxtuyQigZC7EyVg9WHqAW4aMo4QJV/Dw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-06T21:49:08.104781Z","bundle_sha256":"ce9b3d00038d92795d62278bc28d2fac38e9bd2f5069afbbc3b178101a104b00"}}