{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:BIUKUU7ZQANMQBC4P2MBWOEF23","short_pith_number":"pith:BIUKUU7Z","schema_version":"1.0","canonical_sha256":"0a28aa53f9801ac8045c7e981b3885d6f6bf34ce25027b4cfee574b0e79b9a96","source":{"kind":"arxiv","id":"2608.04964","version":1},"attestation_state":"computed","paper":{"title":"WorldCycle: Self-Verifiable Reinforcement Learning for Long-Horizon Video World Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Alan Zhao, Bohai Gu, Dazhao Du, Haobin Zhong, Jian Liu, Jie Zhang, Song Guo, Taiyi Wu, Xiaocheng Lu, Xiaotong Zhao, Xiaoyi Pang, Yueyang Yuan","submitted_at":"2026-08-05T15:34:47Z","abstract_excerpt":"Interactive video world models are essential for long-horizon planning and exploration, yet they suffer from compounding errors. Post-training methods such as reinforcement learning (RL) can improve these models, but they hit a verification bottleneck: for arbitrary action sequences, no ground-truth future state exists to measure long-term drift. Our key insight is that reversible action cycles make this verification possible: a sequence composed with its inverse must analytically return to the initial state, yielding annotation-free supervision on long-horizon correctness. Building on this, w"},"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":"2608.04964","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-08-05T15:34:47Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"925d66ad20cdd406da1dd1c1fbf00fd6f19d4b857aa1fe72eb2977d88972eb11","abstract_canon_sha256":"c8e6c7c521b3ba0a4173d7bf3bd18fb8f02b3b963baed9db676af08766f32a90"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-06T01:47:57.647213Z","signature_b64":"Md7Ygz1GgjpEUXG2Mg7vuxVrfO6CkQQskbFXN5McwDOd7y/QNSZBC8x8ifOaHORElgkoSRHhPb7honKecYwiAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0a28aa53f9801ac8045c7e981b3885d6f6bf34ce25027b4cfee574b0e79b9a96","last_reissued_at":"2026-08-06T01:47:57.645857Z","signature_status":"signed_v1","first_computed_at":"2026-08-06T01:47:57.645857Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"WorldCycle: Self-Verifiable Reinforcement Learning for Long-Horizon Video World Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Alan Zhao, Bohai Gu, Dazhao Du, Haobin Zhong, Jian Liu, Jie Zhang, Song Guo, Taiyi Wu, Xiaocheng Lu, Xiaotong Zhao, Xiaoyi Pang, Yueyang Yuan","submitted_at":"2026-08-05T15:34:47Z","abstract_excerpt":"Interactive video world models are essential for long-horizon planning and exploration, yet they suffer from compounding errors. Post-training methods such as reinforcement learning (RL) can improve these models, but they hit a verification bottleneck: for arbitrary action sequences, no ground-truth future state exists to measure long-term drift. Our key insight is that reversible action cycles make this verification possible: a sequence composed with its inverse must analytically return to the initial state, yielding annotation-free supervision on long-horizon correctness. Building on this, w"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.04964","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/2608.04964/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":"2608.04964","created_at":"2026-08-06T01:47:57.647655+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.04964v1","created_at":"2026-08-06T01:47:57.647655+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.04964","created_at":"2026-08-06T01:47:57.647655+00:00"},{"alias_kind":"pith_short_12","alias_value":"BIUKUU7ZQANM","created_at":"2026-08-06T01:47:57.647655+00:00"},{"alias_kind":"pith_short_16","alias_value":"BIUKUU7ZQANMQBC4","created_at":"2026-08-06T01:47:57.647655+00:00"},{"alias_kind":"pith_short_8","alias_value":"BIUKUU7Z","created_at":"2026-08-06T01:47:57.647655+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/BIUKUU7ZQANMQBC4P2MBWOEF23","json":"https://pith.science/pith/BIUKUU7ZQANMQBC4P2MBWOEF23.json","graph_json":"https://pith.science/api/pith-number/BIUKUU7ZQANMQBC4P2MBWOEF23/graph.json","events_json":"https://pith.science/api/pith-number/BIUKUU7ZQANMQBC4P2MBWOEF23/events.json","paper":"https://pith.science/paper/BIUKUU7Z"},"agent_actions":{"view_html":"https://pith.science/pith/BIUKUU7ZQANMQBC4P2MBWOEF23","download_json":"https://pith.science/pith/BIUKUU7ZQANMQBC4P2MBWOEF23.json","view_paper":"https://pith.science/paper/BIUKUU7Z","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.04964&json=true","fetch_graph":"https://pith.science/api/pith-number/BIUKUU7ZQANMQBC4P2MBWOEF23/graph.json","fetch_events":"https://pith.science/api/pith-number/BIUKUU7ZQANMQBC4P2MBWOEF23/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BIUKUU7ZQANMQBC4P2MBWOEF23/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BIUKUU7ZQANMQBC4P2MBWOEF23/action/storage_attestation","attest_author":"https://pith.science/pith/BIUKUU7ZQANMQBC4P2MBWOEF23/action/author_attestation","sign_citation":"https://pith.science/pith/BIUKUU7ZQANMQBC4P2MBWOEF23/action/citation_signature","submit_replication":"https://pith.science/pith/BIUKUU7ZQANMQBC4P2MBWOEF23/action/replication_record"}},"created_at":"2026-08-06T01:47:57.647655+00:00","updated_at":"2026-08-06T01:47:57.647655+00:00"}