{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:XFWMFNOAHHTJXS5MDZOX2K2CYB","short_pith_number":"pith:XFWMFNOA","schema_version":"1.0","canonical_sha256":"b96cc2b5c039e69bcbac1e5d7d2b42c059ac7b614f5742fd3615a7da664653ca","source":{"kind":"arxiv","id":"2506.16565","version":1},"attestation_state":"computed","paper":{"title":"Reimagination with Test-time Observation Interventions: Distractor-Robust World Model Predictions for Visual Model Predictive Control","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.RO","authors_text":"Andrea Bajcsy, Boyi Li, Chenfeng Xu, Jianglan Wei, Masayoshi Tomizuka, Ran Tian, Yuxin Chen","submitted_at":"2025-06-19T19:41:29Z","abstract_excerpt":"World models enable robots to \"imagine\" future observations given current observations and planned actions, and have been increasingly adopted as generalized dynamics models to facilitate robot learning. Despite their promise, these models remain brittle when encountering novel visual distractors such as objects and background elements rarely seen during training. Specifically, novel distractors can corrupt action outcome predictions, causing downstream failures when robots rely on the world model imaginations for planning or action verification. In this work, we propose Reimagination with Obs"},"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":"2506.16565","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2025-06-19T19:41:29Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"b36250f0dc579e25f1b070e9f05f7e1480d561cf63eeb50bd0c9eb0575f8280c","abstract_canon_sha256":"c25391c5c63523ae6cde5ed3a213cb9a35b62c81d9665a80b799b77355acd332"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:24:38.491028Z","signature_b64":"PfmPq1oobxB9LJ7J45yxpJJ82uHNhCOTQTykrSGWULRYZLfGMrjPJNKbR/qi0Il3g2MTa8CTWDd958vkG/dgBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b96cc2b5c039e69bcbac1e5d7d2b42c059ac7b614f5742fd3615a7da664653ca","last_reissued_at":"2026-07-05T11:24:38.490616Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:24:38.490616Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reimagination with Test-time Observation Interventions: Distractor-Robust World Model Predictions for Visual Model Predictive Control","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.RO","authors_text":"Andrea Bajcsy, Boyi Li, Chenfeng Xu, Jianglan Wei, Masayoshi Tomizuka, Ran Tian, Yuxin Chen","submitted_at":"2025-06-19T19:41:29Z","abstract_excerpt":"World models enable robots to \"imagine\" future observations given current observations and planned actions, and have been increasingly adopted as generalized dynamics models to facilitate robot learning. Despite their promise, these models remain brittle when encountering novel visual distractors such as objects and background elements rarely seen during training. Specifically, novel distractors can corrupt action outcome predictions, causing downstream failures when robots rely on the world model imaginations for planning or action verification. In this work, we propose Reimagination with Obs"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.16565","kind":"arxiv","version":1},"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/2506.16565/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":"2506.16565","created_at":"2026-07-05T11:24:38.490669+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.16565v1","created_at":"2026-07-05T11:24:38.490669+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.16565","created_at":"2026-07-05T11:24:38.490669+00:00"},{"alias_kind":"pith_short_12","alias_value":"XFWMFNOAHHTJ","created_at":"2026-07-05T11:24:38.490669+00:00"},{"alias_kind":"pith_short_16","alias_value":"XFWMFNOAHHTJXS5M","created_at":"2026-07-05T11:24:38.490669+00:00"},{"alias_kind":"pith_short_8","alias_value":"XFWMFNOA","created_at":"2026-07-05T11:24:38.490669+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.22998","citing_title":"TEXEDO : Test Time Scaling for Controller-aware Language-conditioned Humanoid Motion Generation","ref_index":3,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XFWMFNOAHHTJXS5MDZOX2K2CYB","json":"https://pith.science/pith/XFWMFNOAHHTJXS5MDZOX2K2CYB.json","graph_json":"https://pith.science/api/pith-number/XFWMFNOAHHTJXS5MDZOX2K2CYB/graph.json","events_json":"https://pith.science/api/pith-number/XFWMFNOAHHTJXS5MDZOX2K2CYB/events.json","paper":"https://pith.science/paper/XFWMFNOA"},"agent_actions":{"view_html":"https://pith.science/pith/XFWMFNOAHHTJXS5MDZOX2K2CYB","download_json":"https://pith.science/pith/XFWMFNOAHHTJXS5MDZOX2K2CYB.json","view_paper":"https://pith.science/paper/XFWMFNOA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.16565&json=true","fetch_graph":"https://pith.science/api/pith-number/XFWMFNOAHHTJXS5MDZOX2K2CYB/graph.json","fetch_events":"https://pith.science/api/pith-number/XFWMFNOAHHTJXS5MDZOX2K2CYB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XFWMFNOAHHTJXS5MDZOX2K2CYB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XFWMFNOAHHTJXS5MDZOX2K2CYB/action/storage_attestation","attest_author":"https://pith.science/pith/XFWMFNOAHHTJXS5MDZOX2K2CYB/action/author_attestation","sign_citation":"https://pith.science/pith/XFWMFNOAHHTJXS5MDZOX2K2CYB/action/citation_signature","submit_replication":"https://pith.science/pith/XFWMFNOAHHTJXS5MDZOX2K2CYB/action/replication_record"}},"created_at":"2026-07-05T11:24:38.490669+00:00","updated_at":"2026-07-05T11:24:38.490669+00:00"}