{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:OHIG3E7ZSCI4CLW5RNOKDU2MXZ","short_pith_number":"pith:OHIG3E7Z","schema_version":"1.0","canonical_sha256":"71d06d93f99091c12edd8b5ca1d34cbe7dfbd1b9cd427b9336a830fb191c2552","source":{"kind":"arxiv","id":"2608.06706","version":1},"attestation_state":"computed","paper":{"title":"Dueling World Models: Advantage-Style Action Channels for Common-Mode Distractor Rejection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Heikichi Hayashi, Jiazhuo Li, Yiming Fei, Zhiruo Zhou","submitted_at":"2026-08-07T02:02:19Z","abstract_excerpt":"Latent world models plan by predicting future states from an action, but when a scene contains motion the agent does not control, they quietly go action-blind: predictions for different actions become indistinguishable even as the training loss keeps improving. Existing remedies suppress this distraction with reconstruction, task reward, or auxiliary objectives, each adding machinery or assumptions. We show that a minimal alternative suffices, borrowed from the dueling decomposition of value into a state baseline and an action advantage: in latent dynamics, subtracting a prediction's mean effe"},"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":"2608.06706","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-08-07T02:02:19Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"da031f114e45b74eb122f524e74f8e7291282f25ae6c4b5c0ea8d443c1ccdf61","abstract_canon_sha256":"5f75446bcd55d888f639dc1ec31ecca215470a192d6f8448c5d7cab9347c179a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-10T01:11:46.177181Z","signature_b64":"20cGC7Ovo649Hd/mfbZllEd505HEYn4+OdzgfqjzyoQ1r9Pk/uAAL7xyOVTWhx/D4XV7h65bx9wxbcd/6tKWAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"71d06d93f99091c12edd8b5ca1d34cbe7dfbd1b9cd427b9336a830fb191c2552","last_reissued_at":"2026-08-10T01:11:46.174708Z","signature_status":"signed_v1","first_computed_at":"2026-08-10T01:11:46.174708Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Dueling World Models: Advantage-Style Action Channels for Common-Mode Distractor Rejection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Heikichi Hayashi, Jiazhuo Li, Yiming Fei, Zhiruo Zhou","submitted_at":"2026-08-07T02:02:19Z","abstract_excerpt":"Latent world models plan by predicting future states from an action, but when a scene contains motion the agent does not control, they quietly go action-blind: predictions for different actions become indistinguishable even as the training loss keeps improving. Existing remedies suppress this distraction with reconstruction, task reward, or auxiliary objectives, each adding machinery or assumptions. We show that a minimal alternative suffices, borrowed from the dueling decomposition of value into a state baseline and an action advantage: in latent dynamics, subtracting a prediction's mean effe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.06706","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/2608.06706/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":"2608.06706","created_at":"2026-08-10T01:11:46.175789+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.06706v1","created_at":"2026-08-10T01:11:46.175789+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.06706","created_at":"2026-08-10T01:11:46.175789+00:00"},{"alias_kind":"pith_short_12","alias_value":"OHIG3E7ZSCI4","created_at":"2026-08-10T01:11:46.175789+00:00"},{"alias_kind":"pith_short_16","alias_value":"OHIG3E7ZSCI4CLW5","created_at":"2026-08-10T01:11:46.175789+00:00"},{"alias_kind":"pith_short_8","alias_value":"OHIG3E7Z","created_at":"2026-08-10T01:11:46.175789+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/OHIG3E7ZSCI4CLW5RNOKDU2MXZ","json":"https://pith.science/pith/OHIG3E7ZSCI4CLW5RNOKDU2MXZ.json","graph_json":"https://pith.science/api/pith-number/OHIG3E7ZSCI4CLW5RNOKDU2MXZ/graph.json","events_json":"https://pith.science/api/pith-number/OHIG3E7ZSCI4CLW5RNOKDU2MXZ/events.json","paper":"https://pith.science/paper/OHIG3E7Z"},"agent_actions":{"view_html":"https://pith.science/pith/OHIG3E7ZSCI4CLW5RNOKDU2MXZ","download_json":"https://pith.science/pith/OHIG3E7ZSCI4CLW5RNOKDU2MXZ.json","view_paper":"https://pith.science/paper/OHIG3E7Z","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.06706&json=true","fetch_graph":"https://pith.science/api/pith-number/OHIG3E7ZSCI4CLW5RNOKDU2MXZ/graph.json","fetch_events":"https://pith.science/api/pith-number/OHIG3E7ZSCI4CLW5RNOKDU2MXZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OHIG3E7ZSCI4CLW5RNOKDU2MXZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OHIG3E7ZSCI4CLW5RNOKDU2MXZ/action/storage_attestation","attest_author":"https://pith.science/pith/OHIG3E7ZSCI4CLW5RNOKDU2MXZ/action/author_attestation","sign_citation":"https://pith.science/pith/OHIG3E7ZSCI4CLW5RNOKDU2MXZ/action/citation_signature","submit_replication":"https://pith.science/pith/OHIG3E7ZSCI4CLW5RNOKDU2MXZ/action/replication_record"}},"created_at":"2026-08-10T01:11:46.175789+00:00","updated_at":"2026-08-10T01:11:46.175789+00:00"}