{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:PLSN4G3AY2R3LTTGDUX3Y5YEGG","short_pith_number":"pith:PLSN4G3A","schema_version":"1.0","canonical_sha256":"7ae4de1b60c6a3b5ce661d2fbc77043187d17a8d96e8dd5c7d981a07bf1fb536","source":{"kind":"arxiv","id":"2404.06356","version":1},"attestation_state":"computed","paper":{"title":"Policy-Guided Diffusion","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.RO"],"primary_cat":"cs.LG","authors_text":"Benjamin Ellis, Cong Lu, Jakob Foerster, Matthew Thomas Jackson, Michael Tryfan Matthews, Shimon Whiteson","submitted_at":"2024-04-09T14:46:48Z","abstract_excerpt":"In many real-world settings, agents must learn from an offline dataset gathered by some prior behavior policy. Such a setting naturally leads to distribution shift between the behavior policy and the target policy being trained - requiring policy conservatism to avoid instability and overestimation bias. Autoregressive world models offer a different solution to this by generating synthetic, on-policy experience. However, in practice, model rollouts must be severely truncated to avoid compounding error. As an alternative, we propose policy-guided diffusion. Our method uses diffusion models to g"},"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":"2404.06356","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-04-09T14:46:48Z","cross_cats_sorted":["cs.AI","cs.RO"],"title_canon_sha256":"67a3a9f9564ce404ac12db6471e2ae1cadebf183f80d3ee7d17dd30a43b64e01","abstract_canon_sha256":"8145822fecc2c1f67c4539cef3a4841c0353159f2704c009aaad322bf249bf3f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:06:08.077511Z","signature_b64":"wHEuqJ/KoWZDzOV95cXt7SgJLnOEBE9zSAftWPsVZkEoGjb5wNcg/Dg4FPV5kA3H2VqotsxJCUbO9id7bsZnDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7ae4de1b60c6a3b5ce661d2fbc77043187d17a8d96e8dd5c7d981a07bf1fb536","last_reissued_at":"2026-07-05T08:06:08.077019Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:06:08.077019Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Policy-Guided Diffusion","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.RO"],"primary_cat":"cs.LG","authors_text":"Benjamin Ellis, Cong Lu, Jakob Foerster, Matthew Thomas Jackson, Michael Tryfan Matthews, Shimon Whiteson","submitted_at":"2024-04-09T14:46:48Z","abstract_excerpt":"In many real-world settings, agents must learn from an offline dataset gathered by some prior behavior policy. Such a setting naturally leads to distribution shift between the behavior policy and the target policy being trained - requiring policy conservatism to avoid instability and overestimation bias. Autoregressive world models offer a different solution to this by generating synthetic, on-policy experience. However, in practice, model rollouts must be severely truncated to avoid compounding error. As an alternative, we propose policy-guided diffusion. Our method uses diffusion models to g"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.06356","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/2404.06356/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":"2404.06356","created_at":"2026-07-05T08:06:08.077080+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.06356v1","created_at":"2026-07-05T08:06:08.077080+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.06356","created_at":"2026-07-05T08:06:08.077080+00:00"},{"alias_kind":"pith_short_12","alias_value":"PLSN4G3AY2R3","created_at":"2026-07-05T08:06:08.077080+00:00"},{"alias_kind":"pith_short_16","alias_value":"PLSN4G3AY2R3LTTG","created_at":"2026-07-05T08:06:08.077080+00:00"},{"alias_kind":"pith_short_8","alias_value":"PLSN4G3A","created_at":"2026-07-05T08:06:08.077080+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":9,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.05558","citing_title":"Autoregressive Diffusion World Models for Off-Policy Evaluation of LLM Agents","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00336","citing_title":"From Noise to Control: Parameterized Diffusion Policies","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24810","citing_title":"Cross-Domain Energy-Guided Diffusion Generation for Off-Dynamics Reinforcement Learning","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2506.05762","citing_title":"BiTrajDiff: Bidirectional Trajectory Generation with Diffusion Models for Offline Reinforcement Learning","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2509.19538","citing_title":"DAWM: Diffusion Action World Models for Offline Reinforcement Learning via Action-Inferred Transitions","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2409.00588","citing_title":"Diffusion Policy Policy Optimization","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13013","citing_title":"JEDI: Joint Embedding Diffusion World Model for Online Model-Based Reinforcement Learning","ref_index":63,"is_internal_anchor":false},{"citing_arxiv_id":"2605.03075","citing_title":"Refining Compositional Diffusion for Reliable Long-Horizon Planning","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2604.09035","citing_title":"Advantage-Guided Diffusion for Model-Based Reinforcement Learning","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PLSN4G3AY2R3LTTGDUX3Y5YEGG","json":"https://pith.science/pith/PLSN4G3AY2R3LTTGDUX3Y5YEGG.json","graph_json":"https://pith.science/api/pith-number/PLSN4G3AY2R3LTTGDUX3Y5YEGG/graph.json","events_json":"https://pith.science/api/pith-number/PLSN4G3AY2R3LTTGDUX3Y5YEGG/events.json","paper":"https://pith.science/paper/PLSN4G3A"},"agent_actions":{"view_html":"https://pith.science/pith/PLSN4G3AY2R3LTTGDUX3Y5YEGG","download_json":"https://pith.science/pith/PLSN4G3AY2R3LTTGDUX3Y5YEGG.json","view_paper":"https://pith.science/paper/PLSN4G3A","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.06356&json=true","fetch_graph":"https://pith.science/api/pith-number/PLSN4G3AY2R3LTTGDUX3Y5YEGG/graph.json","fetch_events":"https://pith.science/api/pith-number/PLSN4G3AY2R3LTTGDUX3Y5YEGG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PLSN4G3AY2R3LTTGDUX3Y5YEGG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PLSN4G3AY2R3LTTGDUX3Y5YEGG/action/storage_attestation","attest_author":"https://pith.science/pith/PLSN4G3AY2R3LTTGDUX3Y5YEGG/action/author_attestation","sign_citation":"https://pith.science/pith/PLSN4G3AY2R3LTTGDUX3Y5YEGG/action/citation_signature","submit_replication":"https://pith.science/pith/PLSN4G3AY2R3LTTGDUX3Y5YEGG/action/replication_record"}},"created_at":"2026-07-05T08:06:08.077080+00:00","updated_at":"2026-07-05T08:06:08.077080+00:00"}