{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:S463HN2IFPWM7OWUUY3SYN4VPA","short_pith_number":"pith:S463HN2I","schema_version":"1.0","canonical_sha256":"973db3b7482beccfbad4a6372c3795783a4d65744b267ab9d203278ee0d57214","source":{"kind":"arxiv","id":"2310.13024","version":1},"attestation_state":"computed","paper":{"title":"Towards Anytime Fine-tuning: Continually Pre-trained Language Models with Hypernetwork Prompt","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Caigao Jiang, Defu Lian, Gangwei Jiang, James Y. Zhang, Jun Zhou, Siqiao Xue, Ying Wei","submitted_at":"2023-10-19T06:34:40Z","abstract_excerpt":"Continual pre-training has been urgent for adapting a pre-trained model to a multitude of domains and tasks in the fast-evolving world. In practice, a continually pre-trained model is expected to demonstrate not only greater capacity when fine-tuned on pre-trained domains but also a non-decreasing performance on unseen ones. In this work, we first investigate such anytime fine-tuning effectiveness of existing continual pre-training approaches, concluding with unanimously decreased performance on unseen domains. To this end, we propose a prompt-guided continual pre-training method, where we tra"},"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":"2310.13024","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-19T06:34:40Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"1bd8cd4234d675909ed11b92d79cfa69e413dd1503f97bb0e19800d29f15708d","abstract_canon_sha256":"c11f1103529121045d59e0a63b288bf2b9a62e07cc1cae69d70c9ea2b7be0f13"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:02:55.968193Z","signature_b64":"szPHllYzoAvOcQbivI2zIsUnRjP/4AyEe1XdTNxhWsel0iUoaYx7Jlm5/fuqw1b8aljdJjwws9l8HaIA0GQDCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"973db3b7482beccfbad4a6372c3795783a4d65744b267ab9d203278ee0d57214","last_reissued_at":"2026-07-05T07:02:55.967704Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:02:55.967704Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards Anytime Fine-tuning: Continually Pre-trained Language Models with Hypernetwork Prompt","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Caigao Jiang, Defu Lian, Gangwei Jiang, James Y. Zhang, Jun Zhou, Siqiao Xue, Ying Wei","submitted_at":"2023-10-19T06:34:40Z","abstract_excerpt":"Continual pre-training has been urgent for adapting a pre-trained model to a multitude of domains and tasks in the fast-evolving world. In practice, a continually pre-trained model is expected to demonstrate not only greater capacity when fine-tuned on pre-trained domains but also a non-decreasing performance on unseen ones. In this work, we first investigate such anytime fine-tuning effectiveness of existing continual pre-training approaches, concluding with unanimously decreased performance on unseen domains. To this end, we propose a prompt-guided continual pre-training method, where we tra"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.13024","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/2310.13024/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":"2310.13024","created_at":"2026-07-05T07:02:55.967765+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.13024v1","created_at":"2026-07-05T07:02:55.967765+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.13024","created_at":"2026-07-05T07:02:55.967765+00:00"},{"alias_kind":"pith_short_12","alias_value":"S463HN2IFPWM","created_at":"2026-07-05T07:02:55.967765+00:00"},{"alias_kind":"pith_short_16","alias_value":"S463HN2IFPWM7OWU","created_at":"2026-07-05T07:02:55.967765+00:00"},{"alias_kind":"pith_short_8","alias_value":"S463HN2I","created_at":"2026-07-05T07:02:55.967765+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.14725","citing_title":"GRID: Scaling Task-Agnostic Inference in Continual Prompt Tuning","ref_index":15,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/S463HN2IFPWM7OWUUY3SYN4VPA","json":"https://pith.science/pith/S463HN2IFPWM7OWUUY3SYN4VPA.json","graph_json":"https://pith.science/api/pith-number/S463HN2IFPWM7OWUUY3SYN4VPA/graph.json","events_json":"https://pith.science/api/pith-number/S463HN2IFPWM7OWUUY3SYN4VPA/events.json","paper":"https://pith.science/paper/S463HN2I"},"agent_actions":{"view_html":"https://pith.science/pith/S463HN2IFPWM7OWUUY3SYN4VPA","download_json":"https://pith.science/pith/S463HN2IFPWM7OWUUY3SYN4VPA.json","view_paper":"https://pith.science/paper/S463HN2I","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.13024&json=true","fetch_graph":"https://pith.science/api/pith-number/S463HN2IFPWM7OWUUY3SYN4VPA/graph.json","fetch_events":"https://pith.science/api/pith-number/S463HN2IFPWM7OWUUY3SYN4VPA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/S463HN2IFPWM7OWUUY3SYN4VPA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/S463HN2IFPWM7OWUUY3SYN4VPA/action/storage_attestation","attest_author":"https://pith.science/pith/S463HN2IFPWM7OWUUY3SYN4VPA/action/author_attestation","sign_citation":"https://pith.science/pith/S463HN2IFPWM7OWUUY3SYN4VPA/action/citation_signature","submit_replication":"https://pith.science/pith/S463HN2IFPWM7OWUUY3SYN4VPA/action/replication_record"}},"created_at":"2026-07-05T07:02:55.967765+00:00","updated_at":"2026-07-05T07:02:55.967765+00:00"}