{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:36ODUBJHSKAULXG34VDRA2EEDF","short_pith_number":"pith:36ODUBJH","schema_version":"1.0","canonical_sha256":"df9c3a0527928145dcdbe5471068841946d12b568eb97961bd8499bf4839bbeb","source":{"kind":"arxiv","id":"2602.07399","version":2},"attestation_state":"computed","paper":{"title":"VGAS: Value-Guided Action-Chunk Selection for Few-Shot Vision-Language-Action Adaptation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.AI","authors_text":"Changhua Xu, En Yu, Jie Lu, Junyu Xuan","submitted_at":"2026-02-07T06:31:53Z","abstract_excerpt":"Vision--Language--Action (VLA) models bridge multimodal reasoning with physical control, but adapting them to new tasks with scarce demonstrations remains unreliable. While fine-tuned VLA policies often produce semantically plausible trajectories, failures often arise from unresolved geometric ambiguities, where near-miss actions lead to divergent execution outcomes under limited supervision. We study few-shot VLA adaptation from a \\emph{generation--selection} perspective and propose a novel framework \\textbf{VGAS} (\\textbf{V}alue-\\textbf{G}uided \\textbf{A}ction-chunk \\textbf{S}election). It p"},"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":"2602.07399","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-02-07T06:31:53Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"503c6c36bdce9a5eb704ec76aa11d981cd324f771cdefbd9bef0a980f3301d58","abstract_canon_sha256":"1be7ce8dc9cfbadc41c9f90a0b3d0a08dac40b84784230086fab29c370198f72"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-25T02:01:14.699807Z","signature_b64":"BNj1HB2tAwVNkJ6OYHY0WNAMXFcfA11fm/etkY6zNTINlKDvyEOOh/uyjhpr3Z3PR0kpMgUmXbmZqmqC/ubDDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"df9c3a0527928145dcdbe5471068841946d12b568eb97961bd8499bf4839bbeb","last_reissued_at":"2026-05-25T02:01:14.699081Z","signature_status":"signed_v1","first_computed_at":"2026-05-25T02:01:14.699081Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VGAS: Value-Guided Action-Chunk Selection for Few-Shot Vision-Language-Action Adaptation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.AI","authors_text":"Changhua Xu, En Yu, Jie Lu, Junyu Xuan","submitted_at":"2026-02-07T06:31:53Z","abstract_excerpt":"Vision--Language--Action (VLA) models bridge multimodal reasoning with physical control, but adapting them to new tasks with scarce demonstrations remains unreliable. While fine-tuned VLA policies often produce semantically plausible trajectories, failures often arise from unresolved geometric ambiguities, where near-miss actions lead to divergent execution outcomes under limited supervision. We study few-shot VLA adaptation from a \\emph{generation--selection} perspective and propose a novel framework \\textbf{VGAS} (\\textbf{V}alue-\\textbf{G}uided \\textbf{A}ction-chunk \\textbf{S}election). It p"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.07399","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/2602.07399/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":"2602.07399","created_at":"2026-05-25T02:01:14.699194+00:00"},{"alias_kind":"arxiv_version","alias_value":"2602.07399v2","created_at":"2026-05-25T02:01:14.699194+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.07399","created_at":"2026-05-25T02:01:14.699194+00:00"},{"alias_kind":"pith_short_12","alias_value":"36ODUBJHSKAU","created_at":"2026-05-25T02:01:14.699194+00:00"},{"alias_kind":"pith_short_16","alias_value":"36ODUBJHSKAULXG3","created_at":"2026-05-25T02:01:14.699194+00:00"},{"alias_kind":"pith_short_8","alias_value":"36ODUBJH","created_at":"2026-05-25T02:01:14.699194+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.01295","citing_title":"Autonomous Drift Learning in Data Streams: A Unified Perspective","ref_index":194,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/36ODUBJHSKAULXG34VDRA2EEDF","json":"https://pith.science/pith/36ODUBJHSKAULXG34VDRA2EEDF.json","graph_json":"https://pith.science/api/pith-number/36ODUBJHSKAULXG34VDRA2EEDF/graph.json","events_json":"https://pith.science/api/pith-number/36ODUBJHSKAULXG34VDRA2EEDF/events.json","paper":"https://pith.science/paper/36ODUBJH"},"agent_actions":{"view_html":"https://pith.science/pith/36ODUBJHSKAULXG34VDRA2EEDF","download_json":"https://pith.science/pith/36ODUBJHSKAULXG34VDRA2EEDF.json","view_paper":"https://pith.science/paper/36ODUBJH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2602.07399&json=true","fetch_graph":"https://pith.science/api/pith-number/36ODUBJHSKAULXG34VDRA2EEDF/graph.json","fetch_events":"https://pith.science/api/pith-number/36ODUBJHSKAULXG34VDRA2EEDF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/36ODUBJHSKAULXG34VDRA2EEDF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/36ODUBJHSKAULXG34VDRA2EEDF/action/storage_attestation","attest_author":"https://pith.science/pith/36ODUBJHSKAULXG34VDRA2EEDF/action/author_attestation","sign_citation":"https://pith.science/pith/36ODUBJHSKAULXG34VDRA2EEDF/action/citation_signature","submit_replication":"https://pith.science/pith/36ODUBJHSKAULXG34VDRA2EEDF/action/replication_record"}},"created_at":"2026-05-25T02:01:14.699194+00:00","updated_at":"2026-05-25T02:01:14.699194+00:00"}