{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:AWP4YKGS72BW3GXMW4LNZSIXJ3","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"c3e76177c1ad4fb5620334875c4f8b76877e9f0d120abfebce4e269ab89f436c","cross_cats_sorted":["cs.LG","math.PR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2024-10-02T05:14:28Z","title_canon_sha256":"27d556e01d8b157f9908df21ad78bd166c06bcf13b9cc3feca19063bc210afd2"},"schema_version":"1.0","source":{"id":"2410.01244","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.01244","created_at":"2026-06-30T01:17:20Z"},{"alias_kind":"arxiv_version","alias_value":"2410.01244v2","created_at":"2026-06-30T01:17:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.01244","created_at":"2026-06-30T01:17:20Z"},{"alias_kind":"pith_short_12","alias_value":"AWP4YKGS72BW","created_at":"2026-06-30T01:17:20Z"},{"alias_kind":"pith_short_16","alias_value":"AWP4YKGS72BW3GXM","created_at":"2026-06-30T01:17:20Z"},{"alias_kind":"pith_short_8","alias_value":"AWP4YKGS","created_at":"2026-06-30T01:17:20Z"}],"graph_snapshots":[{"event_id":"sha256:21b0f7f382b504cc27b878453baaa207cf3b3b8942f71b3b04e5141a1bac925e","target":"graph","created_at":"2026-06-30T01:17:20Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"one can learn the score of a symmetrized distribution using equivariant vector fields without data augmentations through the analysis of the optimality and equivalence of score-matching objectives. This also provides practical guidance that one does not have to augment the dataset as long as the vector field or the neural network parametrization is equivariant."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The data distribution is exactly group-invariant and the symmetry group is known in advance so that an exactly equivariant vector field can be constructed; the improved d1 bound and the HJB equivalence both rest on this premise (abstract, paragraphs on improved d1 bound and HJB analysis)."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Equivariant SGMs achieve improved Wasserstein-1 generalization bounds on group-invariant distributions and learn the symmetrized score via equivariant vector fields without augmentation, with non-equivariant models incurring a quantifiable model-form error."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Equivariant vector fields enable score-based generative models to learn group-invariant distributions without data augmentation."}],"snapshot_sha256":"19435ceb3f2896af078439d40ac2ef8cdc9a1aa65f748ef88ed5d0e7273c95c7"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2410.01244/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce a novel Wasserstein-1 ($W_1$) path-space divergence for stochastic and deterministic dynamics and establish a Wasserstein Uncertainty Propagation (WUP) theorem that bounds the $W_1$ distance between terminal distributions by the proposed divergence, equivalently characterized by a weighted $L^2$ discrepancy between the underlying drifts and the $W_1$ distance between their initial measures. A key ingredient is a probabilistic framework combining adjoint Feynman-Kac representations with synchronous coupling (and reflection coupling on bounded domains), yielding Wasserstein stabilit","authors_text":"Benjamin J. Zhang, Markos A. Katsoulakis, Ziyu Chen","cross_cats":["cs.LG","math.PR"],"headline":"Equivariant vector fields enable score-based generative models to learn group-invariant distributions without data augmentation.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2024-10-02T05:14:28Z","title":"Robustness and Structure Preservation in Flow-Based Generative Models via Wasserstein Path-Space Divergences"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.01244","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-23T20:31:36.267911Z","id":"ea2aa5ea-9787-4cf2-887f-a63dfdbb2fb5","model_set":{"reader":"grok-4.3"},"one_line_summary":"Equivariant SGMs achieve improved Wasserstein-1 generalization bounds on group-invariant distributions and learn the symmetrized score via equivariant vector fields without augmentation, with non-equivariant models incurring a quantifiable model-form error.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Equivariant vector fields enable score-based generative models to learn group-invariant distributions without data augmentation.","strongest_claim":"one can learn the score of a symmetrized distribution using equivariant vector fields without data augmentations through the analysis of the optimality and equivalence of score-matching objectives. This also provides practical guidance that one does not have to augment the dataset as long as the vector field or the neural network parametrization is equivariant.","weakest_assumption":"The data distribution is exactly group-invariant and the symmetry group is known in advance so that an exactly equivariant vector field can be constructed; the improved d1 bound and the HJB equivalence both rest on this premise (abstract, paragraphs on improved d1 bound and HJB analysis)."}},"verdict_id":"ea2aa5ea-9787-4cf2-887f-a63dfdbb2fb5"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:ced5fb7e5474bbc3782380f9c66cf8d2a05eaa559792069774f1ffb13b684fe2","target":"record","created_at":"2026-06-30T01:17:20Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"c3e76177c1ad4fb5620334875c4f8b76877e9f0d120abfebce4e269ab89f436c","cross_cats_sorted":["cs.LG","math.PR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2024-10-02T05:14:28Z","title_canon_sha256":"27d556e01d8b157f9908df21ad78bd166c06bcf13b9cc3feca19063bc210afd2"},"schema_version":"1.0","source":{"id":"2410.01244","kind":"arxiv","version":2}},"canonical_sha256":"059fcc28d2fe836d9aecb716dcc9174ecd3eed40b3a918822a2cb013d0b41f92","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"059fcc28d2fe836d9aecb716dcc9174ecd3eed40b3a918822a2cb013d0b41f92","first_computed_at":"2026-06-30T01:17:20.567791Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-30T01:17:20.567791Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"CmxxnZ9GQy2sYSAPeo+JhdHg6OIOYUnaJ0PNBJsAl6DU4Mu9FCZoIc47R9Iz7oct1BfC6mKuTzXyODbicc74Bw==","signature_status":"signed_v1","signed_at":"2026-06-30T01:17:20.568339Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.01244","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ced5fb7e5474bbc3782380f9c66cf8d2a05eaa559792069774f1ffb13b684fe2","sha256:21b0f7f382b504cc27b878453baaa207cf3b3b8942f71b3b04e5141a1bac925e"],"state_sha256":"c88d9888de220140e3161cd2bbc3ecd92ac1bca6d6d3ce075ebaad1f0d2756cb"}