{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:2L5LNCUBWNSPGVCFKLY2BOICAA","short_pith_number":"pith:2L5LNCUB","schema_version":"1.0","canonical_sha256":"d2fab68a81b364f3544552f1a0b902000b9e76752bc5ff83d6e08ca093baf727","source":{"kind":"arxiv","id":"2203.06654","version":1},"attestation_state":"computed","paper":{"title":"Continual Prompt Tuning for Dialog State Tracking","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bing Li, Fei Mi, Minlie Huang, Qi Zhu, Xiaoyan Zhu","submitted_at":"2022-03-13T13:22:41Z","abstract_excerpt":"A desirable dialog system should be able to continually learn new skills without forgetting old ones, and thereby adapt to new domains or tasks in its life cycle. However, continually training a model often leads to a well-known catastrophic forgetting issue. In this paper, we present Continual Prompt Tuning, a parameter-efficient framework that not only avoids forgetting but also enables knowledge transfer between tasks. To avoid forgetting, we only learn and store a few prompt tokens' embeddings for each task while freezing the backbone pre-trained model. To achieve bi-directional knowledge "},"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":"2203.06654","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-03-13T13:22:41Z","cross_cats_sorted":[],"title_canon_sha256":"f5400205a3886fa89f1b57b32df95b099812d20118484faeddf522a566c95ec7","abstract_canon_sha256":"e6fd67667e5490a1f651fedbc9fbc642266b8429e13eea2303418fe23b8ec526"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:04:53.329264Z","signature_b64":"I1c+ArP/6xe+SzpAsvYsZY6mFMG6Y0nukXuMvsZs5hVIxxs9ems5TA85zImnoCeasFSdlQR2VU+SMbTL/x5DDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d2fab68a81b364f3544552f1a0b902000b9e76752bc5ff83d6e08ca093baf727","last_reissued_at":"2026-07-05T04:04:53.328844Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:04:53.328844Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Continual Prompt Tuning for Dialog State Tracking","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bing Li, Fei Mi, Minlie Huang, Qi Zhu, Xiaoyan Zhu","submitted_at":"2022-03-13T13:22:41Z","abstract_excerpt":"A desirable dialog system should be able to continually learn new skills without forgetting old ones, and thereby adapt to new domains or tasks in its life cycle. However, continually training a model often leads to a well-known catastrophic forgetting issue. In this paper, we present Continual Prompt Tuning, a parameter-efficient framework that not only avoids forgetting but also enables knowledge transfer between tasks. To avoid forgetting, we only learn and store a few prompt tokens' embeddings for each task while freezing the backbone pre-trained model. To achieve bi-directional knowledge "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.06654","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/2203.06654/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":"2203.06654","created_at":"2026-07-05T04:04:53.328902+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.06654v1","created_at":"2026-07-05T04:04:53.328902+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.06654","created_at":"2026-07-05T04:04:53.328902+00:00"},{"alias_kind":"pith_short_12","alias_value":"2L5LNCUBWNSP","created_at":"2026-07-05T04:04:53.328902+00:00"},{"alias_kind":"pith_short_16","alias_value":"2L5LNCUBWNSPGVCF","created_at":"2026-07-05T04:04:53.328902+00:00"},{"alias_kind":"pith_short_8","alias_value":"2L5LNCUB","created_at":"2026-07-05T04:04:53.328902+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2403.14608","citing_title":"Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Survey","ref_index":172,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2L5LNCUBWNSPGVCFKLY2BOICAA","json":"https://pith.science/pith/2L5LNCUBWNSPGVCFKLY2BOICAA.json","graph_json":"https://pith.science/api/pith-number/2L5LNCUBWNSPGVCFKLY2BOICAA/graph.json","events_json":"https://pith.science/api/pith-number/2L5LNCUBWNSPGVCFKLY2BOICAA/events.json","paper":"https://pith.science/paper/2L5LNCUB"},"agent_actions":{"view_html":"https://pith.science/pith/2L5LNCUBWNSPGVCFKLY2BOICAA","download_json":"https://pith.science/pith/2L5LNCUBWNSPGVCFKLY2BOICAA.json","view_paper":"https://pith.science/paper/2L5LNCUB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.06654&json=true","fetch_graph":"https://pith.science/api/pith-number/2L5LNCUBWNSPGVCFKLY2BOICAA/graph.json","fetch_events":"https://pith.science/api/pith-number/2L5LNCUBWNSPGVCFKLY2BOICAA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2L5LNCUBWNSPGVCFKLY2BOICAA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2L5LNCUBWNSPGVCFKLY2BOICAA/action/storage_attestation","attest_author":"https://pith.science/pith/2L5LNCUBWNSPGVCFKLY2BOICAA/action/author_attestation","sign_citation":"https://pith.science/pith/2L5LNCUBWNSPGVCFKLY2BOICAA/action/citation_signature","submit_replication":"https://pith.science/pith/2L5LNCUBWNSPGVCFKLY2BOICAA/action/replication_record"}},"created_at":"2026-07-05T04:04:53.328902+00:00","updated_at":"2026-07-05T04:04:53.328902+00:00"}