{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:YGKQNGPWBV43DAYFJP5ZLOTVIG","short_pith_number":"pith:YGKQNGPW","schema_version":"1.0","canonical_sha256":"c1950699f60d79b183054bfb95ba7541bf45758fc19cb86fda9fd5aeb9ced8d8","source":{"kind":"arxiv","id":"2012.15283","version":3},"attestation_state":"computed","paper":{"title":"ECONET: Effective Continual Pretraining of Language Models for Event Temporal Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Nanyun Peng, Rujun Han, Xiang Ren","submitted_at":"2020-12-30T18:57:16Z","abstract_excerpt":"While pre-trained language models (PTLMs) have achieved noticeable success on many NLP tasks, they still struggle for tasks that require event temporal reasoning, which is essential for event-centric applications. We present a continual pre-training approach that equips PTLMs with targeted knowledge about event temporal relations. We design self-supervised learning objectives to recover masked-out event and temporal indicators and to discriminate sentences from their corrupted counterparts (where event or temporal indicators got replaced). By further pre-training a PTLM with these objectives j"},"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":"2012.15283","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-12-30T18:57:16Z","cross_cats_sorted":[],"title_canon_sha256":"302ec519ebd1037bfbeafad2618fc3c7d93bedac615d1aed2d3f2ccb50572400","abstract_canon_sha256":"1a75df6943575306977361b6cb314c9fb6bf5f4ec2b83b494f6ec883ae9d164d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:15:08.569239Z","signature_b64":"a2F7RjL7siChRfgXiDqMNMEzfzhvtWU1XyRQs/74YNb8SI3rCYD6gGK7QRZ5HlZ0SpcdyZ9lqXHY9XKD7wv/Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c1950699f60d79b183054bfb95ba7541bf45758fc19cb86fda9fd5aeb9ced8d8","last_reissued_at":"2026-07-05T03:15:08.568905Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:15:08.568905Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ECONET: Effective Continual Pretraining of Language Models for Event Temporal Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Nanyun Peng, Rujun Han, Xiang Ren","submitted_at":"2020-12-30T18:57:16Z","abstract_excerpt":"While pre-trained language models (PTLMs) have achieved noticeable success on many NLP tasks, they still struggle for tasks that require event temporal reasoning, which is essential for event-centric applications. We present a continual pre-training approach that equips PTLMs with targeted knowledge about event temporal relations. We design self-supervised learning objectives to recover masked-out event and temporal indicators and to discriminate sentences from their corrupted counterparts (where event or temporal indicators got replaced). By further pre-training a PTLM with these objectives j"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2012.15283","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/2012.15283/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":"2012.15283","created_at":"2026-07-05T03:15:08.568959+00:00"},{"alias_kind":"arxiv_version","alias_value":"2012.15283v3","created_at":"2026-07-05T03:15:08.568959+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2012.15283","created_at":"2026-07-05T03:15:08.568959+00:00"},{"alias_kind":"pith_short_12","alias_value":"YGKQNGPWBV43","created_at":"2026-07-05T03:15:08.568959+00:00"},{"alias_kind":"pith_short_16","alias_value":"YGKQNGPWBV43DAYF","created_at":"2026-07-05T03:15:08.568959+00:00"},{"alias_kind":"pith_short_8","alias_value":"YGKQNGPW","created_at":"2026-07-05T03:15:08.568959+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.13052","citing_title":"RAG-Enhanced Large Language Models for Dynamic Content Expiration Prediction in Web Search","ref_index":4,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YGKQNGPWBV43DAYFJP5ZLOTVIG","json":"https://pith.science/pith/YGKQNGPWBV43DAYFJP5ZLOTVIG.json","graph_json":"https://pith.science/api/pith-number/YGKQNGPWBV43DAYFJP5ZLOTVIG/graph.json","events_json":"https://pith.science/api/pith-number/YGKQNGPWBV43DAYFJP5ZLOTVIG/events.json","paper":"https://pith.science/paper/YGKQNGPW"},"agent_actions":{"view_html":"https://pith.science/pith/YGKQNGPWBV43DAYFJP5ZLOTVIG","download_json":"https://pith.science/pith/YGKQNGPWBV43DAYFJP5ZLOTVIG.json","view_paper":"https://pith.science/paper/YGKQNGPW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2012.15283&json=true","fetch_graph":"https://pith.science/api/pith-number/YGKQNGPWBV43DAYFJP5ZLOTVIG/graph.json","fetch_events":"https://pith.science/api/pith-number/YGKQNGPWBV43DAYFJP5ZLOTVIG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YGKQNGPWBV43DAYFJP5ZLOTVIG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YGKQNGPWBV43DAYFJP5ZLOTVIG/action/storage_attestation","attest_author":"https://pith.science/pith/YGKQNGPWBV43DAYFJP5ZLOTVIG/action/author_attestation","sign_citation":"https://pith.science/pith/YGKQNGPWBV43DAYFJP5ZLOTVIG/action/citation_signature","submit_replication":"https://pith.science/pith/YGKQNGPWBV43DAYFJP5ZLOTVIG/action/replication_record"}},"created_at":"2026-07-05T03:15:08.568959+00:00","updated_at":"2026-07-05T03:15:08.568959+00:00"}