{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:33YXYH6VUYQUBIFL4GH7JATHRF","short_pith_number":"pith:33YXYH6V","schema_version":"1.0","canonical_sha256":"def17c1fd5a62140a0abe18ff4826789553a7ee02f7b80933f49f2655bd0c556","source":{"kind":"arxiv","id":"2512.02581","version":3},"attestation_state":"computed","paper":{"title":"Training Diffusion Policies via Prior-Mapping Co-Evolution","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bo An, Chubin Zhang, Feng Chen, Fuchao Yang, Ivor Tsang, Lang Feng, Xingrui Yu, Yang You, Yaxin Zhou, Zhenglin Wan","submitted_at":"2025-12-02T09:49:26Z","abstract_excerpt":"Reinforcement learning (RL) faces a persistent tension: policies that are stable to optimize (e.g., Gaussians) are often too simple to represent the multimodal action distributions required for complex control. Conversely, expressive generative policies -- such as diffusion and flow matching -- can be difficult to optimize in online RL due to intractable likelihoods and gradients propagating through long sampling chains. We address this tension with a key structural principle: decoupling optimization from generation. Building on this, we introduce GoRL (Generative Online Reinforcement Learning"},"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":"2512.02581","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-12-02T09:49:26Z","cross_cats_sorted":[],"title_canon_sha256":"10fc9199b84ac41b3970de659faa40036c0ee39fdf4ede49e6111721aa8547b0","abstract_canon_sha256":"fe91732a2c660308121d3ddaccd5427e7c482338cc108707791cbd973cb2e1eb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T01:12:50.260164Z","signature_b64":"MSBJDes1v8XxIkDxJbO8+NAf7xOiT+oPyh1BCfCfpE6iBNmSSY5ar+zWjKQjml6oJlZCoP4cLV3jJKtn3LETCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"def17c1fd5a62140a0abe18ff4826789553a7ee02f7b80933f49f2655bd0c556","last_reissued_at":"2026-06-23T01:12:50.259601Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T01:12:50.259601Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Training Diffusion Policies via Prior-Mapping Co-Evolution","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bo An, Chubin Zhang, Feng Chen, Fuchao Yang, Ivor Tsang, Lang Feng, Xingrui Yu, Yang You, Yaxin Zhou, Zhenglin Wan","submitted_at":"2025-12-02T09:49:26Z","abstract_excerpt":"Reinforcement learning (RL) faces a persistent tension: policies that are stable to optimize (e.g., Gaussians) are often too simple to represent the multimodal action distributions required for complex control. Conversely, expressive generative policies -- such as diffusion and flow matching -- can be difficult to optimize in online RL due to intractable likelihoods and gradients propagating through long sampling chains. We address this tension with a key structural principle: decoupling optimization from generation. Building on this, we introduce GoRL (Generative Online Reinforcement Learning"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2512.02581","kind":"arxiv","version":3},"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/2512.02581/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":"2512.02581","created_at":"2026-06-23T01:12:50.259671+00:00"},{"alias_kind":"arxiv_version","alias_value":"2512.02581v3","created_at":"2026-06-23T01:12:50.259671+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2512.02581","created_at":"2026-06-23T01:12:50.259671+00:00"},{"alias_kind":"pith_short_12","alias_value":"33YXYH6VUYQU","created_at":"2026-06-23T01:12:50.259671+00:00"},{"alias_kind":"pith_short_16","alias_value":"33YXYH6VUYQUBIFL","created_at":"2026-06-23T01:12:50.259671+00:00"},{"alias_kind":"pith_short_8","alias_value":"33YXYH6V","created_at":"2026-06-23T01:12:50.259671+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/33YXYH6VUYQUBIFL4GH7JATHRF","json":"https://pith.science/pith/33YXYH6VUYQUBIFL4GH7JATHRF.json","graph_json":"https://pith.science/api/pith-number/33YXYH6VUYQUBIFL4GH7JATHRF/graph.json","events_json":"https://pith.science/api/pith-number/33YXYH6VUYQUBIFL4GH7JATHRF/events.json","paper":"https://pith.science/paper/33YXYH6V"},"agent_actions":{"view_html":"https://pith.science/pith/33YXYH6VUYQUBIFL4GH7JATHRF","download_json":"https://pith.science/pith/33YXYH6VUYQUBIFL4GH7JATHRF.json","view_paper":"https://pith.science/paper/33YXYH6V","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2512.02581&json=true","fetch_graph":"https://pith.science/api/pith-number/33YXYH6VUYQUBIFL4GH7JATHRF/graph.json","fetch_events":"https://pith.science/api/pith-number/33YXYH6VUYQUBIFL4GH7JATHRF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/33YXYH6VUYQUBIFL4GH7JATHRF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/33YXYH6VUYQUBIFL4GH7JATHRF/action/storage_attestation","attest_author":"https://pith.science/pith/33YXYH6VUYQUBIFL4GH7JATHRF/action/author_attestation","sign_citation":"https://pith.science/pith/33YXYH6VUYQUBIFL4GH7JATHRF/action/citation_signature","submit_replication":"https://pith.science/pith/33YXYH6VUYQUBIFL4GH7JATHRF/action/replication_record"}},"created_at":"2026-06-23T01:12:50.259671+00:00","updated_at":"2026-06-23T01:12:50.259671+00:00"}