{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:OFGGA7DHWHS7DIUNJBJKI5TI7H","short_pith_number":"pith:OFGGA7DH","schema_version":"1.0","canonical_sha256":"714c607c67b1e5f1a28d4852a47668f9d087d8b1a78f06590bb431ab2c7d7a24","source":{"kind":"arxiv","id":"2601.09512","version":2},"attestation_state":"computed","paper":{"title":"CLARE: Continual Learning for Vision-Language-Action Models via Autonomous Adapter Routing and Expansion","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.RO","authors_text":"Angela P. Schoellig, Ralf R\\\"omer, Yi Zhang, Yuming Li","submitted_at":"2026-01-14T14:23:42Z","abstract_excerpt":"To teach robots complex manipulation tasks, a common approach is to fine-tune a pre-trained vision-language-action model (VLA) on task-specific data. However, since this recipe updates existing representations, it is unsuitable for long-term operation in the real world, where robots must continually adapt to new tasks and environments while retaining the knowledge they have already acquired. Existing continual learning methods for robotics commonly require storing previous data (exemplars), struggle with long task sequences, or rely on task identifiers for deployment. To address these limitati"},"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":"2601.09512","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2026-01-14T14:23:42Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"7804b82cd1a576bdab5ff55d7451a268a31d1aac3ecd3b55fecd77bbc74b4d52","abstract_canon_sha256":"476ec7603502b63ed2346ff0ab34dcf205d17acda6ca3d7d198e2bb50fccc2b5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:00:32.094987Z","signature_b64":"k4M/NfdlfXenM/4mqGVL1YgRjQs3y7cK0idC2DQFwnh59hDdDAce3BUWK0DMQHHTvrjLwoprUQAXg8bB3V8CBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"714c607c67b1e5f1a28d4852a47668f9d087d8b1a78f06590bb431ab2c7d7a24","last_reissued_at":"2026-05-20T00:00:32.094507Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:00:32.094507Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CLARE: Continual Learning for Vision-Language-Action Models via Autonomous Adapter Routing and Expansion","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.RO","authors_text":"Angela P. Schoellig, Ralf R\\\"omer, Yi Zhang, Yuming Li","submitted_at":"2026-01-14T14:23:42Z","abstract_excerpt":"To teach robots complex manipulation tasks, a common approach is to fine-tune a pre-trained vision-language-action model (VLA) on task-specific data. However, since this recipe updates existing representations, it is unsuitable for long-term operation in the real world, where robots must continually adapt to new tasks and environments while retaining the knowledge they have already acquired. Existing continual learning methods for robotics commonly require storing previous data (exemplars), struggle with long task sequences, or rely on task identifiers for deployment. To address these limitati"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2601.09512","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/2601.09512/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":"2601.09512","created_at":"2026-05-20T00:00:32.094581+00:00"},{"alias_kind":"arxiv_version","alias_value":"2601.09512v2","created_at":"2026-05-20T00:00:32.094581+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.09512","created_at":"2026-05-20T00:00:32.094581+00:00"},{"alias_kind":"pith_short_12","alias_value":"OFGGA7DHWHS7","created_at":"2026-05-20T00:00:32.094581+00:00"},{"alias_kind":"pith_short_16","alias_value":"OFGGA7DHWHS7DIUN","created_at":"2026-05-20T00:00:32.094581+00:00"},{"alias_kind":"pith_short_8","alias_value":"OFGGA7DH","created_at":"2026-05-20T00:00:32.094581+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":3,"sample":[{"citing_arxiv_id":"2605.12369","citing_title":"GuidedVLA: Specifying Task-Relevant Factors via Plug-and-Play Action Attention Specialization","ref_index":74,"is_internal_anchor":true},{"citing_arxiv_id":"2605.07381","citing_title":"Escaping the Diversity Trap in Robotic Manipulation via Anchor-Centric Adaptation","ref_index":44,"is_internal_anchor":true},{"citing_arxiv_id":"2604.14375","citing_title":"Modular Continual Learning via Zero-Leakage Reconstruction Routing and Autonomous Task Discovery","ref_index":22,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OFGGA7DHWHS7DIUNJBJKI5TI7H","json":"https://pith.science/pith/OFGGA7DHWHS7DIUNJBJKI5TI7H.json","graph_json":"https://pith.science/api/pith-number/OFGGA7DHWHS7DIUNJBJKI5TI7H/graph.json","events_json":"https://pith.science/api/pith-number/OFGGA7DHWHS7DIUNJBJKI5TI7H/events.json","paper":"https://pith.science/paper/OFGGA7DH"},"agent_actions":{"view_html":"https://pith.science/pith/OFGGA7DHWHS7DIUNJBJKI5TI7H","download_json":"https://pith.science/pith/OFGGA7DHWHS7DIUNJBJKI5TI7H.json","view_paper":"https://pith.science/paper/OFGGA7DH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2601.09512&json=true","fetch_graph":"https://pith.science/api/pith-number/OFGGA7DHWHS7DIUNJBJKI5TI7H/graph.json","fetch_events":"https://pith.science/api/pith-number/OFGGA7DHWHS7DIUNJBJKI5TI7H/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OFGGA7DHWHS7DIUNJBJKI5TI7H/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OFGGA7DHWHS7DIUNJBJKI5TI7H/action/storage_attestation","attest_author":"https://pith.science/pith/OFGGA7DHWHS7DIUNJBJKI5TI7H/action/author_attestation","sign_citation":"https://pith.science/pith/OFGGA7DHWHS7DIUNJBJKI5TI7H/action/citation_signature","submit_replication":"https://pith.science/pith/OFGGA7DHWHS7DIUNJBJKI5TI7H/action/replication_record"}},"created_at":"2026-05-20T00:00:32.094581+00:00","updated_at":"2026-05-20T00:00:32.094581+00:00"}