{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:ZARD27V76RNBZ5IFD7F5FHBUYV","short_pith_number":"pith:ZARD27V7","schema_version":"1.0","canonical_sha256":"c8223d7ebff45a1cf5051fcbd29c34c55f934fd57294adb660f11ef5352b8c3b","source":{"kind":"arxiv","id":"2602.03668","version":3},"attestation_state":"computed","paper":{"title":"MVP-LAM: Learning Action-Centric Latent Action via Cross-Viewpoint Reconstruction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Cross-viewpoint reconstruction trains latent actions to capture underlying robot actions rather than camera-specific details.","cross_cats":["cs.CV"],"primary_cat":"cs.RO","authors_text":"Dohyeok Lee, Jin Woo Koo, Jung Min Lee, Jungwoo Lee, Li Zhao, Sangwoo Hong, Seokhun Ju, Taehyun Cho","submitted_at":"2026-02-03T15:51:25Z","abstract_excerpt":"Latent actions learned from diverse human videos serve as pseudo-labels for vision-language-action (VLA) pretraining, but provide effective supervision only if they remain informative about the underlying ground-truth actions. For effective supervision, latent actions should contain information about the underlying actions even though they are inaccessible. We propose Multi-ViewPoint Latent Action Moel (MVP-LAM), which learns latent actions that are highly informative about ground-truth actions from multi-view videos. MVP-LAM trains latent actions with a cross-viewpoint reconstruction objectiv"},"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":true},"canonical_record":{"source":{"id":"2602.03668","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2026-02-03T15:51:25Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"dadabbac91cc2e0b0c79aceb634cafd2e709dfa00da201181bf898701319c8a7","abstract_canon_sha256":"45e4aaeaa709cd485cc2eb7893f6e4469d4d6f62a8ed77de72c0efa5046b2dd0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-28T01:04:36.264470Z","signature_b64":"sdJ9AvZXezzT778XkE01+YVKGCmeGsVjnXMM2TbwF2z4wp6ogSZ1c4S1k4ZnyeRzlbQ+4Vr3VY1ae37vZ42dBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c8223d7ebff45a1cf5051fcbd29c34c55f934fd57294adb660f11ef5352b8c3b","last_reissued_at":"2026-05-28T01:04:36.264068Z","signature_status":"signed_v1","first_computed_at":"2026-05-28T01:04:36.264068Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MVP-LAM: Learning Action-Centric Latent Action via Cross-Viewpoint Reconstruction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Cross-viewpoint reconstruction trains latent actions to capture underlying robot actions rather than camera-specific details.","cross_cats":["cs.CV"],"primary_cat":"cs.RO","authors_text":"Dohyeok Lee, Jin Woo Koo, Jung Min Lee, Jungwoo Lee, Li Zhao, Sangwoo Hong, Seokhun Ju, Taehyun Cho","submitted_at":"2026-02-03T15:51:25Z","abstract_excerpt":"Latent actions learned from diverse human videos serve as pseudo-labels for vision-language-action (VLA) pretraining, but provide effective supervision only if they remain informative about the underlying ground-truth actions. For effective supervision, latent actions should contain information about the underlying actions even though they are inaccessible. We propose Multi-ViewPoint Latent Action Moel (MVP-LAM), which learns latent actions that are highly informative about ground-truth actions from multi-view videos. MVP-LAM trains latent actions with a cross-viewpoint reconstruction objectiv"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"MVP-LAM produces more action-centric latent actions, achieving higher mutual information with ground-truth actions and improved action prediction, including under out-of-distribution evaluation. Finally, pretraining VLAs with MVP-LAM latent actions improves downstream manipulation performance on various benchmarks.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That forcing a latent action from one viewpoint to explain the future in another viewpoint will make the latent action contain information about the underlying ground-truth actions rather than viewpoint-specific cues.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"MVP-LAM learns action-centric latent actions from multi-view videos via cross-viewpoint reconstruction, yielding higher mutual information with ground-truth actions and improved downstream VLA manipulation performance.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Cross-viewpoint reconstruction trains latent actions to capture underlying robot actions rather than camera-specific details.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"e9f85e4e3c684adf15bd89c51595407fc5614ba562c36a381fa0d59d78391f3d"},"source":{"id":"2602.03668","kind":"arxiv","version":3},"verdict":{"id":"4791e08e-34dd-42c2-a1ad-5cfcd5a82804","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T07:47:36.715161Z","strongest_claim":"MVP-LAM produces more action-centric latent actions, achieving higher mutual information with ground-truth actions and improved action prediction, including under out-of-distribution evaluation. Finally, pretraining VLAs with MVP-LAM latent actions improves downstream manipulation performance on various benchmarks.","one_line_summary":"MVP-LAM learns action-centric latent actions from multi-view videos via cross-viewpoint reconstruction, yielding higher mutual information with ground-truth actions and improved downstream VLA manipulation performance.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That forcing a latent action from one viewpoint to explain the future in another viewpoint will make the latent action contain information about the underlying ground-truth actions rather than viewpoint-specific cues.","pith_extraction_headline":"Cross-viewpoint reconstruction trains latent actions to capture underlying robot actions rather than camera-specific details."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2602.03668/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":2,"snapshot_sha256":"73afbef700494a2e36d6b5812d9ab7e286420d4f39f91637ba681de9fddad6f1"},"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":"2602.03668","created_at":"2026-05-28T01:04:36.264123+00:00"},{"alias_kind":"arxiv_version","alias_value":"2602.03668v3","created_at":"2026-05-28T01:04:36.264123+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.03668","created_at":"2026-05-28T01:04:36.264123+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZARD27V76RNB","created_at":"2026-05-28T01:04:36.264123+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZARD27V76RNBZ5IF","created_at":"2026-05-28T01:04:36.264123+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZARD27V7","created_at":"2026-05-28T01:04:36.264123+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.20223","citing_title":"Why Latent Actions Fail, and How to Prevent It","ref_index":5,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZARD27V76RNBZ5IFD7F5FHBUYV","json":"https://pith.science/pith/ZARD27V76RNBZ5IFD7F5FHBUYV.json","graph_json":"https://pith.science/api/pith-number/ZARD27V76RNBZ5IFD7F5FHBUYV/graph.json","events_json":"https://pith.science/api/pith-number/ZARD27V76RNBZ5IFD7F5FHBUYV/events.json","paper":"https://pith.science/paper/ZARD27V7"},"agent_actions":{"view_html":"https://pith.science/pith/ZARD27V76RNBZ5IFD7F5FHBUYV","download_json":"https://pith.science/pith/ZARD27V76RNBZ5IFD7F5FHBUYV.json","view_paper":"https://pith.science/paper/ZARD27V7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2602.03668&json=true","fetch_graph":"https://pith.science/api/pith-number/ZARD27V76RNBZ5IFD7F5FHBUYV/graph.json","fetch_events":"https://pith.science/api/pith-number/ZARD27V76RNBZ5IFD7F5FHBUYV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZARD27V76RNBZ5IFD7F5FHBUYV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZARD27V76RNBZ5IFD7F5FHBUYV/action/storage_attestation","attest_author":"https://pith.science/pith/ZARD27V76RNBZ5IFD7F5FHBUYV/action/author_attestation","sign_citation":"https://pith.science/pith/ZARD27V76RNBZ5IFD7F5FHBUYV/action/citation_signature","submit_replication":"https://pith.science/pith/ZARD27V76RNBZ5IFD7F5FHBUYV/action/replication_record"}},"created_at":"2026-05-28T01:04:36.264123+00:00","updated_at":"2026-05-28T01:04:36.264123+00:00"}