{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:XNZALFT5O7MJTFL5GXC7NUDRGR","short_pith_number":"pith:XNZALFT5","schema_version":"1.0","canonical_sha256":"bb7205967d77d899957d35c5f6d07134711739c2292da165c797876650e7484d","source":{"kind":"arxiv","id":"2109.04912","version":1},"attestation_state":"computed","paper":{"title":"ReasonBERT: Pre-trained to Reason with Distant Supervision","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Alyssa Lees, Cong Yu, Huan Sun, Xiang Deng, You Wu, Yu Su","submitted_at":"2021-09-10T14:49:44Z","abstract_excerpt":"We present ReasonBert, a pre-training method that augments language models with the ability to reason over long-range relations and multiple, possibly hybrid contexts. Unlike existing pre-training methods that only harvest learning signals from local contexts of naturally occurring texts, we propose a generalized notion of distant supervision to automatically connect multiple pieces of text and tables to create pre-training examples that require long-range reasoning. Different types of reasoning are simulated, including intersecting multiple pieces of evidence, bridging from one piece of evide"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2109.04912","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-09-10T14:49:44Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"b307b9ed3f02cfe38cf126340d09699d36132fe3ef3af48cf9c1ef0f63cbcc7f","abstract_canon_sha256":"1337a79ee87b248d4175cd5bd71c384307aeee787322586b7214babb740a43e4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:13:11.410196Z","signature_b64":"Wn38HTz3XyAe3ugtbhPiPj/pavheGcmTOou78KyAStB5pnzWYpV/X61Q6sQYfA5foWWvcLbbaHjcIIEE3GvlBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bb7205967d77d899957d35c5f6d07134711739c2292da165c797876650e7484d","last_reissued_at":"2026-07-05T03:13:11.409723Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:13:11.409723Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ReasonBERT: Pre-trained to Reason with Distant Supervision","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Alyssa Lees, Cong Yu, Huan Sun, Xiang Deng, You Wu, Yu Su","submitted_at":"2021-09-10T14:49:44Z","abstract_excerpt":"We present ReasonBert, a pre-training method that augments language models with the ability to reason over long-range relations and multiple, possibly hybrid contexts. Unlike existing pre-training methods that only harvest learning signals from local contexts of naturally occurring texts, we propose a generalized notion of distant supervision to automatically connect multiple pieces of text and tables to create pre-training examples that require long-range reasoning. Different types of reasoning are simulated, including intersecting multiple pieces of evidence, bridging from one piece of evide"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2109.04912","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/2109.04912/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2109.04912","created_at":"2026-07-05T03:13:11.409780+00:00"},{"alias_kind":"arxiv_version","alias_value":"2109.04912v1","created_at":"2026-07-05T03:13:11.409780+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2109.04912","created_at":"2026-07-05T03:13:11.409780+00:00"},{"alias_kind":"pith_short_12","alias_value":"XNZALFT5O7MJ","created_at":"2026-07-05T03:13:11.409780+00:00"},{"alias_kind":"pith_short_16","alias_value":"XNZALFT5O7MJTFL5","created_at":"2026-07-05T03:13:11.409780+00:00"},{"alias_kind":"pith_short_8","alias_value":"XNZALFT5","created_at":"2026-07-05T03:13:11.409780+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2207.14255","citing_title":"Efficient Training of Language Models to Fill in the Middle","ref_index":104,"is_internal_anchor":false},{"citing_arxiv_id":"2501.09686","citing_title":"Towards Large Reasoning Models: A Survey of Reinforced Reasoning with Large Language Models","ref_index":29,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XNZALFT5O7MJTFL5GXC7NUDRGR","json":"https://pith.science/pith/XNZALFT5O7MJTFL5GXC7NUDRGR.json","graph_json":"https://pith.science/api/pith-number/XNZALFT5O7MJTFL5GXC7NUDRGR/graph.json","events_json":"https://pith.science/api/pith-number/XNZALFT5O7MJTFL5GXC7NUDRGR/events.json","paper":"https://pith.science/paper/XNZALFT5"},"agent_actions":{"view_html":"https://pith.science/pith/XNZALFT5O7MJTFL5GXC7NUDRGR","download_json":"https://pith.science/pith/XNZALFT5O7MJTFL5GXC7NUDRGR.json","view_paper":"https://pith.science/paper/XNZALFT5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2109.04912&json=true","fetch_graph":"https://pith.science/api/pith-number/XNZALFT5O7MJTFL5GXC7NUDRGR/graph.json","fetch_events":"https://pith.science/api/pith-number/XNZALFT5O7MJTFL5GXC7NUDRGR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XNZALFT5O7MJTFL5GXC7NUDRGR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XNZALFT5O7MJTFL5GXC7NUDRGR/action/storage_attestation","attest_author":"https://pith.science/pith/XNZALFT5O7MJTFL5GXC7NUDRGR/action/author_attestation","sign_citation":"https://pith.science/pith/XNZALFT5O7MJTFL5GXC7NUDRGR/action/citation_signature","submit_replication":"https://pith.science/pith/XNZALFT5O7MJTFL5GXC7NUDRGR/action/replication_record"}},"created_at":"2026-07-05T03:13:11.409780+00:00","updated_at":"2026-07-05T03:13:11.409780+00:00"}