{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:TGTLWJTBFRLTEXNVEEKFPPOCYI","short_pith_number":"pith:TGTLWJTB","schema_version":"1.0","canonical_sha256":"99a6bb26612c57325db5211457bdc2c22829da21838d32282bad9beebc364dcc","source":{"kind":"arxiv","id":"2312.11752","version":5},"attestation_state":"computed","paper":{"title":"Learning a Diffusion Model Policy from Rewards via Q-Score Matching","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Alejandro Escontrela, Michael Psenka, Pieter Abbeel, Yi Ma","submitted_at":"2023-12-18T23:31:01Z","abstract_excerpt":"Diffusion models have become a popular choice for representing actor policies in behavior cloning and offline reinforcement learning. This is due to their natural ability to optimize an expressive class of distributions over a continuous space. However, previous works fail to exploit the score-based structure of diffusion models, and instead utilize a simple behavior cloning term to train the actor, limiting their ability in the actor-critic setting. In this paper, we present a theoretical framework linking the structure of diffusion model policies to a learned Q-function, by linking the struc"},"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":"2312.11752","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-12-18T23:31:01Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"418e4aba633311c87da4e07a729aa12f0453c9eafb56c5fdd2954e43c232d6d1","abstract_canon_sha256":"a5b2380e5441225390647836d9e33c5c0b36082000c3f766304870cacd6810e4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:14:02.310468Z","signature_b64":"wHRGkWcJ+e4rSSVYnjv2v2eUlBDv+eaFIAsJ0aBTFfEHVk/EjL/LTnpGrwA+/8dqDY8LdHihyY/+bjH7ADV4Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"99a6bb26612c57325db5211457bdc2c22829da21838d32282bad9beebc364dcc","last_reissued_at":"2026-07-05T10:14:02.309973Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:14:02.309973Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning a Diffusion Model Policy from Rewards via Q-Score Matching","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Alejandro Escontrela, Michael Psenka, Pieter Abbeel, Yi Ma","submitted_at":"2023-12-18T23:31:01Z","abstract_excerpt":"Diffusion models have become a popular choice for representing actor policies in behavior cloning and offline reinforcement learning. This is due to their natural ability to optimize an expressive class of distributions over a continuous space. However, previous works fail to exploit the score-based structure of diffusion models, and instead utilize a simple behavior cloning term to train the actor, limiting their ability in the actor-critic setting. In this paper, we present a theoretical framework linking the structure of diffusion model policies to a learned Q-function, by linking the struc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.11752","kind":"arxiv","version":5},"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/2312.11752/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":"2312.11752","created_at":"2026-07-05T10:14:02.310032+00:00"},{"alias_kind":"arxiv_version","alias_value":"2312.11752v5","created_at":"2026-07-05T10:14:02.310032+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.11752","created_at":"2026-07-05T10:14:02.310032+00:00"},{"alias_kind":"pith_short_12","alias_value":"TGTLWJTBFRLT","created_at":"2026-07-05T10:14:02.310032+00:00"},{"alias_kind":"pith_short_16","alias_value":"TGTLWJTBFRLTEXNV","created_at":"2026-07-05T10:14:02.310032+00:00"},{"alias_kind":"pith_short_8","alias_value":"TGTLWJTB","created_at":"2026-07-05T10:14:02.310032+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":12,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.11087","citing_title":"Test-Time Gradient Guidance of Flow Policies in Reinforcement Learning","ref_index":53,"is_internal_anchor":false},{"citing_arxiv_id":"2606.06967","citing_title":"GenPO++: Generative Policy Optimization with Jacobian-free Likelihood Ratios","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2605.26282","citing_title":"Scaling World-Model Reinforcement Learning Through Diffusion Policy Optimization","ref_index":46,"is_internal_anchor":false},{"citing_arxiv_id":"2605.27095","citing_title":"Adversarial Dual On-Policy Distillation from Expressive Teacher","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30056","citing_title":"Sample-Efficient Diffusion-based Reinforcement Learning with Critic Guidance","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16520","citing_title":"Global Convergence of Sampling-Based Nonconvex Optimization through Diffusion-Style Smoothing","ref_index":178,"is_internal_anchor":false},{"citing_arxiv_id":"2507.07986","citing_title":"EXPO: Stable Reinforcement Learning with Expressive Policies","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2506.15799","citing_title":"Steering Your Diffusion Policy with Latent Space Reinforcement Learning","ref_index":61,"is_internal_anchor":false},{"citing_arxiv_id":"2409.00588","citing_title":"Diffusion Policy Policy Optimization","ref_index":74,"is_internal_anchor":false},{"citing_arxiv_id":"2602.02924","citing_title":"How Does the Lagrangian Guide Safe Reinforcement Learning through Diffusion Models?","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07101","citing_title":"Decentralized Diffusion Policy Learning for Enhanced Exploration in Cooperative Multi-agent Reinforcement Learning","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14265","citing_title":"Reinforcement Learning via Value Gradient Flow","ref_index":55,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TGTLWJTBFRLTEXNVEEKFPPOCYI","json":"https://pith.science/pith/TGTLWJTBFRLTEXNVEEKFPPOCYI.json","graph_json":"https://pith.science/api/pith-number/TGTLWJTBFRLTEXNVEEKFPPOCYI/graph.json","events_json":"https://pith.science/api/pith-number/TGTLWJTBFRLTEXNVEEKFPPOCYI/events.json","paper":"https://pith.science/paper/TGTLWJTB"},"agent_actions":{"view_html":"https://pith.science/pith/TGTLWJTBFRLTEXNVEEKFPPOCYI","download_json":"https://pith.science/pith/TGTLWJTBFRLTEXNVEEKFPPOCYI.json","view_paper":"https://pith.science/paper/TGTLWJTB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2312.11752&json=true","fetch_graph":"https://pith.science/api/pith-number/TGTLWJTBFRLTEXNVEEKFPPOCYI/graph.json","fetch_events":"https://pith.science/api/pith-number/TGTLWJTBFRLTEXNVEEKFPPOCYI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TGTLWJTBFRLTEXNVEEKFPPOCYI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TGTLWJTBFRLTEXNVEEKFPPOCYI/action/storage_attestation","attest_author":"https://pith.science/pith/TGTLWJTBFRLTEXNVEEKFPPOCYI/action/author_attestation","sign_citation":"https://pith.science/pith/TGTLWJTBFRLTEXNVEEKFPPOCYI/action/citation_signature","submit_replication":"https://pith.science/pith/TGTLWJTBFRLTEXNVEEKFPPOCYI/action/replication_record"}},"created_at":"2026-07-05T10:14:02.310032+00:00","updated_at":"2026-07-05T10:14:02.310032+00:00"}