{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:XKG5HDJH2XG2WSHL3JKFOVEGDJ","short_pith_number":"pith:XKG5HDJH","schema_version":"1.0","canonical_sha256":"ba8dd38d27d5cdab48ebda545754861a5a6572ea559afa25cb6c370b0853c6b5","source":{"kind":"arxiv","id":"2411.16073","version":4},"attestation_state":"computed","paper":{"title":"Soft-TransFormers for Continual Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Task-specific multiplicative masks on attention projections enable continual learning in frozen pre-trained transformers with minimal added parameters.","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Chang D. Yoo, Haeyong Kang","submitted_at":"2024-11-25T03:52:47Z","abstract_excerpt":"Inspired by the Well-initialized Lottery Ticket Hypothesis (WLTH), we introduce Soft-TransFormers (Soft-TF), a continual learning framework that adapts a frozen pre-trained Transformer through task-specific soft subnetworks: real-valued multiplicative masks over the query, key, value, and output projections of selected self-attention layers. The masks are initialized at one, so optimization starts exactly at the pre-trained solution, and mask-space gradient descent is intrinsically biased toward modulating the backbone's dominant pathways; we prove that, under standard convex-Lipschitz assumpt"},"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":"2411.16073","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-11-25T03:52:47Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"e7581f2ed6b79d16e25d45d2b3648fc39fdb50e122bf029079f83f9b51c4a499","abstract_canon_sha256":"f31acb0b0abf2d3e5157b18a9effc475e2a54a029e7a1e40a2411717cd0ba219"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-22T01:22:09.601260Z","signature_b64":"Zh9WitAd6Sk0utKHOV8+zmz2h2DJ3OrZUkJQmhKdxgvXsmCIABgDYq2WoWa0coACkPwkxOBpB8+OSAdAX1TUAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ba8dd38d27d5cdab48ebda545754861a5a6572ea559afa25cb6c370b0853c6b5","last_reissued_at":"2026-07-22T01:22:09.600324Z","signature_status":"signed_v1","first_computed_at":"2026-07-22T01:22:09.600324Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Soft-TransFormers for Continual Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Task-specific multiplicative masks on attention projections enable continual learning in frozen pre-trained transformers with minimal added parameters.","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Chang D. Yoo, Haeyong Kang","submitted_at":"2024-11-25T03:52:47Z","abstract_excerpt":"Inspired by the Well-initialized Lottery Ticket Hypothesis (WLTH), we introduce Soft-TransFormers (Soft-TF), a continual learning framework that adapts a frozen pre-trained Transformer through task-specific soft subnetworks: real-valued multiplicative masks over the query, key, value, and output projections of selected self-attention layers. The masks are initialized at one, so optimization starts exactly at the pre-trained solution, and mask-space gradient descent is intrinsically biased toward modulating the backbone's dominant pathways; we prove that, under standard convex-Lipschitz assumpt"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Across multiple continual learning benchmarks, Soft-TF achieves state-of-the-art performance, consistently outperforming prompt-based, adapter-based, and LoRA-style baselines while requiring minimal additional parameters.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The central claim depends on the premise that task-specific multiplicative masks applied to the key, query, value, and output projections in self-attention, together with a lightweight dual-prompt mechanism, will enable smooth task adaptation while preserving shared representations in the frozen pre-trained transformer.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Soft-TF learns task-specific multiplicative masks on transformer self-attention layers combined with dual prompts to achieve SOTA continual learning performance with minimal added parameters.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Task-specific multiplicative masks on attention projections enable continual learning in frozen pre-trained transformers with minimal added parameters.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"c4ff11323261ab90278de6465da1485fce8de7d6da76c6ef7ef88fdc6cd48316"},"source":{"id":"2411.16073","kind":"arxiv","version":4},"verdict":{"id":"6f61dbe1-c6ba-4b2a-b130-f5d32a313076","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-23T08:11:08.840576Z","strongest_claim":"Across multiple continual learning benchmarks, Soft-TF achieves state-of-the-art performance, consistently outperforming prompt-based, adapter-based, and LoRA-style baselines while requiring minimal additional parameters.","one_line_summary":"Soft-TF learns task-specific multiplicative masks on transformer self-attention layers combined with dual prompts to achieve SOTA continual learning performance with minimal added parameters.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The central claim depends on the premise that task-specific multiplicative masks applied to the key, query, value, and output projections in self-attention, together with a lightweight dual-prompt mechanism, will enable smooth task adaptation while preserving shared representations in the frozen pre-trained transformer.","pith_extraction_headline":"Task-specific multiplicative masks on attention projections enable continual learning in frozen pre-trained transformers with minimal added parameters."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2411.16073/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":"2411.16073","created_at":"2026-07-22T01:22:09.600763+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.16073v4","created_at":"2026-07-22T01:22:09.600763+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.16073","created_at":"2026-07-22T01:22:09.600763+00:00"},{"alias_kind":"pith_short_12","alias_value":"XKG5HDJH2XG2","created_at":"2026-07-22T01:22:09.600763+00:00"},{"alias_kind":"pith_short_16","alias_value":"XKG5HDJH2XG2WSHL","created_at":"2026-07-22T01:22:09.600763+00:00"},{"alias_kind":"pith_short_8","alias_value":"XKG5HDJH","created_at":"2026-07-22T01:22:09.600763+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/XKG5HDJH2XG2WSHL3JKFOVEGDJ","json":"https://pith.science/pith/XKG5HDJH2XG2WSHL3JKFOVEGDJ.json","graph_json":"https://pith.science/api/pith-number/XKG5HDJH2XG2WSHL3JKFOVEGDJ/graph.json","events_json":"https://pith.science/api/pith-number/XKG5HDJH2XG2WSHL3JKFOVEGDJ/events.json","paper":"https://pith.science/paper/XKG5HDJH"},"agent_actions":{"view_html":"https://pith.science/pith/XKG5HDJH2XG2WSHL3JKFOVEGDJ","download_json":"https://pith.science/pith/XKG5HDJH2XG2WSHL3JKFOVEGDJ.json","view_paper":"https://pith.science/paper/XKG5HDJH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.16073&json=true","fetch_graph":"https://pith.science/api/pith-number/XKG5HDJH2XG2WSHL3JKFOVEGDJ/graph.json","fetch_events":"https://pith.science/api/pith-number/XKG5HDJH2XG2WSHL3JKFOVEGDJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XKG5HDJH2XG2WSHL3JKFOVEGDJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XKG5HDJH2XG2WSHL3JKFOVEGDJ/action/storage_attestation","attest_author":"https://pith.science/pith/XKG5HDJH2XG2WSHL3JKFOVEGDJ/action/author_attestation","sign_citation":"https://pith.science/pith/XKG5HDJH2XG2WSHL3JKFOVEGDJ/action/citation_signature","submit_replication":"https://pith.science/pith/XKG5HDJH2XG2WSHL3JKFOVEGDJ/action/replication_record"}},"created_at":"2026-07-22T01:22:09.600763+00:00","updated_at":"2026-07-22T01:22:09.600763+00:00"}