{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:HWVMSN4VYRYK74LDOXJV7RA6AE","short_pith_number":"pith:HWVMSN4V","schema_version":"1.0","canonical_sha256":"3daac93795c470aff16375d35fc41e01376f6569f2608c97262546a6b6809749","source":{"kind":"arxiv","id":"2205.11200","version":2},"attestation_state":"computed","paper":{"title":"BBTv2: Towards a Gradient-Free Future with Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Hong Qian, Tianxiang Sun, Xipeng Qiu, Xuanjing Huang, Yunhua Zhou, Zhengfu He","submitted_at":"2022-05-23T11:10:19Z","abstract_excerpt":"Most downstream adaptation methods tune all or part of the parameters of pre-trained models (PTMs) through gradient descent, where the tuning cost increases linearly with the growth of the model size. By contrast, gradient-free methods only require the forward computation of the PTM to tune the prompt, retaining the benefits of efficient tuning and deployment. Though, past work on gradient-free tuning often introduces gradient descent to seek a good initialization of prompt and lacks versatility across tasks and PTMs. In this paper, we present BBTv2, an improved version of Black-Box Tuning, to"},"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":"2205.11200","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-05-23T11:10:19Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"1db61f987680cc30236973b688683e29f621146c8362df46a2eb1ae6f1fdd59f","abstract_canon_sha256":"d5abbc45adcec38ee131eb68aedb91725d1faea4cc5065dbc4b289ec8bc923c3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:06:31.535706Z","signature_b64":"hngei10cu+ywOxiY5KBs/ZaMNh9E9xuOPTLvMi9Raoe2RIAvqA1QkjmPvsYzOoncMQzydugmMaS6jyXgAqCnAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3daac93795c470aff16375d35fc41e01376f6569f2608c97262546a6b6809749","last_reissued_at":"2026-07-05T05:06:31.535253Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:06:31.535253Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"BBTv2: Towards a Gradient-Free Future with Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Hong Qian, Tianxiang Sun, Xipeng Qiu, Xuanjing Huang, Yunhua Zhou, Zhengfu He","submitted_at":"2022-05-23T11:10:19Z","abstract_excerpt":"Most downstream adaptation methods tune all or part of the parameters of pre-trained models (PTMs) through gradient descent, where the tuning cost increases linearly with the growth of the model size. By contrast, gradient-free methods only require the forward computation of the PTM to tune the prompt, retaining the benefits of efficient tuning and deployment. Though, past work on gradient-free tuning often introduces gradient descent to seek a good initialization of prompt and lacks versatility across tasks and PTMs. In this paper, we present BBTv2, an improved version of Black-Box Tuning, to"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.11200","kind":"arxiv","version":2},"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/2205.11200/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":"2205.11200","created_at":"2026-07-05T05:06:31.535306+00:00"},{"alias_kind":"arxiv_version","alias_value":"2205.11200v2","created_at":"2026-07-05T05:06:31.535306+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.11200","created_at":"2026-07-05T05:06:31.535306+00:00"},{"alias_kind":"pith_short_12","alias_value":"HWVMSN4VYRYK","created_at":"2026-07-05T05:06:31.535306+00:00"},{"alias_kind":"pith_short_16","alias_value":"HWVMSN4VYRYK74LD","created_at":"2026-07-05T05:06:31.535306+00:00"},{"alias_kind":"pith_short_8","alias_value":"HWVMSN4V","created_at":"2026-07-05T05:06:31.535306+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HWVMSN4VYRYK74LDOXJV7RA6AE","json":"https://pith.science/pith/HWVMSN4VYRYK74LDOXJV7RA6AE.json","graph_json":"https://pith.science/api/pith-number/HWVMSN4VYRYK74LDOXJV7RA6AE/graph.json","events_json":"https://pith.science/api/pith-number/HWVMSN4VYRYK74LDOXJV7RA6AE/events.json","paper":"https://pith.science/paper/HWVMSN4V"},"agent_actions":{"view_html":"https://pith.science/pith/HWVMSN4VYRYK74LDOXJV7RA6AE","download_json":"https://pith.science/pith/HWVMSN4VYRYK74LDOXJV7RA6AE.json","view_paper":"https://pith.science/paper/HWVMSN4V","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2205.11200&json=true","fetch_graph":"https://pith.science/api/pith-number/HWVMSN4VYRYK74LDOXJV7RA6AE/graph.json","fetch_events":"https://pith.science/api/pith-number/HWVMSN4VYRYK74LDOXJV7RA6AE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HWVMSN4VYRYK74LDOXJV7RA6AE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HWVMSN4VYRYK74LDOXJV7RA6AE/action/storage_attestation","attest_author":"https://pith.science/pith/HWVMSN4VYRYK74LDOXJV7RA6AE/action/author_attestation","sign_citation":"https://pith.science/pith/HWVMSN4VYRYK74LDOXJV7RA6AE/action/citation_signature","submit_replication":"https://pith.science/pith/HWVMSN4VYRYK74LDOXJV7RA6AE/action/replication_record"}},"created_at":"2026-07-05T05:06:31.535306+00:00","updated_at":"2026-07-05T05:06:31.535306+00:00"}