{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:AIT2U22CLM5TGF7SLLFFPIY6SE","short_pith_number":"pith:AIT2U22C","schema_version":"1.0","canonical_sha256":"0227aa6b425b3b3317f25aca57a31e9117b8bc9e58017000ac369410190564e5","source":{"kind":"arxiv","id":"2605.15603","version":1},"attestation_state":"computed","paper":{"title":"Offline Reinforcement Learning with Universal Horizon Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Hojun Chung, Junseo Lee, Songhwai Oh","submitted_at":"2026-05-15T04:30:30Z","abstract_excerpt":"Model-based reinforcement learning (RL) offers a compelling approach to offline RL by enabling value learning on imagined on-policy trajectories. However, it often suffers from compounding errors due to repeated model inference on self-generated states. While geometric horizon models (GHM) alleviate this issue through direct prediction over a discounted infinite-horizon future, they remain challenged in accurately modeling distant future states. To this end, we introduce universal horizon models (UHM), a generalization of GHM that directly predicts future states under arbitrary horizons. Lever"},"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":"2605.15603","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-15T04:30:30Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"031a9af8f04b7dd64d25d5ad2c70a9380d7cd3144b774cf5a6d83128f46d257c","abstract_canon_sha256":"066b3709245bfd92a7940845fa966068b322d5d4d42657b3da4e47c6578e963c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:01:07.616538Z","signature_b64":"mGKH3ASew7SKQVN1iw8HUCSLsj1KZnQ5gXiR3GuA+ih5HCbs92TE8lRBGjBQGU69JiEvrqGqC+4w3oF+BBahBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0227aa6b425b3b3317f25aca57a31e9117b8bc9e58017000ac369410190564e5","last_reissued_at":"2026-05-20T00:01:07.615561Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:01:07.615561Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Offline Reinforcement Learning with Universal Horizon Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Hojun Chung, Junseo Lee, Songhwai Oh","submitted_at":"2026-05-15T04:30:30Z","abstract_excerpt":"Model-based reinforcement learning (RL) offers a compelling approach to offline RL by enabling value learning on imagined on-policy trajectories. However, it often suffers from compounding errors due to repeated model inference on self-generated states. While geometric horizon models (GHM) alleviate this issue through direct prediction over a discounted infinite-horizon future, they remain challenged in accurately modeling distant future states. To this end, we introduce universal horizon models (UHM), a generalization of GHM that directly predicts future states under arbitrary horizons. Lever"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.15603","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/2605.15603/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-19T19:34:34.894944Z","status":"skipped","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T17:41:56.052786Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"ed997616e8fd7ddcb5f5e3dce9e5a64c8d6b4a2d2310c658f5b19f5b33ae47be"},"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":"2605.15603","created_at":"2026-05-20T00:01:07.615715+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.15603v1","created_at":"2026-05-20T00:01:07.615715+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.15603","created_at":"2026-05-20T00:01:07.615715+00:00"},{"alias_kind":"pith_short_12","alias_value":"AIT2U22CLM5T","created_at":"2026-05-20T00:01:07.615715+00:00"},{"alias_kind":"pith_short_16","alias_value":"AIT2U22CLM5TGF7S","created_at":"2026-05-20T00:01:07.615715+00:00"},{"alias_kind":"pith_short_8","alias_value":"AIT2U22C","created_at":"2026-05-20T00:01:07.615715+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/AIT2U22CLM5TGF7SLLFFPIY6SE","json":"https://pith.science/pith/AIT2U22CLM5TGF7SLLFFPIY6SE.json","graph_json":"https://pith.science/api/pith-number/AIT2U22CLM5TGF7SLLFFPIY6SE/graph.json","events_json":"https://pith.science/api/pith-number/AIT2U22CLM5TGF7SLLFFPIY6SE/events.json","paper":"https://pith.science/paper/AIT2U22C"},"agent_actions":{"view_html":"https://pith.science/pith/AIT2U22CLM5TGF7SLLFFPIY6SE","download_json":"https://pith.science/pith/AIT2U22CLM5TGF7SLLFFPIY6SE.json","view_paper":"https://pith.science/paper/AIT2U22C","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.15603&json=true","fetch_graph":"https://pith.science/api/pith-number/AIT2U22CLM5TGF7SLLFFPIY6SE/graph.json","fetch_events":"https://pith.science/api/pith-number/AIT2U22CLM5TGF7SLLFFPIY6SE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AIT2U22CLM5TGF7SLLFFPIY6SE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AIT2U22CLM5TGF7SLLFFPIY6SE/action/storage_attestation","attest_author":"https://pith.science/pith/AIT2U22CLM5TGF7SLLFFPIY6SE/action/author_attestation","sign_citation":"https://pith.science/pith/AIT2U22CLM5TGF7SLLFFPIY6SE/action/citation_signature","submit_replication":"https://pith.science/pith/AIT2U22CLM5TGF7SLLFFPIY6SE/action/replication_record"}},"created_at":"2026-05-20T00:01:07.615715+00:00","updated_at":"2026-05-20T00:01:07.615715+00:00"}