{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:5JUNLTC42BDTJHVYZZT5AWGIWA","short_pith_number":"pith:5JUNLTC4","schema_version":"1.0","canonical_sha256":"ea68d5cc5cd047349eb8ce67d058c8b01ffe88291f85bc0c0850e334515dbd4e","source":{"kind":"arxiv","id":"2410.02078","version":1},"attestation_state":"computed","paper":{"title":"Posterior sampling via Langevin dynamics based on generative priors","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV","cs.LG"],"primary_cat":"stat.ML","authors_text":"Jianfeng Lu, Matthew Repasky, Qiang Qiu, Vishal Purohit, Xiuyuan Cheng, Yao Xie","submitted_at":"2024-10-02T22:57:47Z","abstract_excerpt":"Posterior sampling in high-dimensional spaces using generative models holds significant promise for various applications, including but not limited to inverse problems and guided generation tasks. Despite many recent developments, generating diverse posterior samples remains a challenge, as existing methods require restarting the entire generative process for each new sample, making the procedure computationally expensive. In this work, we propose efficient posterior sampling by simulating Langevin dynamics in the noise space of a pre-trained generative model. By exploiting the mapping between"},"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":"2410.02078","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2024-10-02T22:57:47Z","cross_cats_sorted":["cs.CV","cs.LG"],"title_canon_sha256":"17ad7330cfedbdf01e4eb0be1b19ad5954661e7202631e9f63f6a7f04b35ed60","abstract_canon_sha256":"06bee9096332594efc36954ec751f838abe16a971a817109295c1c2af8f7738a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:15:00.111432Z","signature_b64":"Qv2Tu8X0Vq9pA+NSmm7gv1ub+mDx5OV+x+0Dn3c1+8KhPRbOPhvTJqvG6IoYEhzkZz78L0GCGyps95c57S9yAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ea68d5cc5cd047349eb8ce67d058c8b01ffe88291f85bc0c0850e334515dbd4e","last_reissued_at":"2026-07-05T09:15:00.110994Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:15:00.110994Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Posterior sampling via Langevin dynamics based on generative priors","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV","cs.LG"],"primary_cat":"stat.ML","authors_text":"Jianfeng Lu, Matthew Repasky, Qiang Qiu, Vishal Purohit, Xiuyuan Cheng, Yao Xie","submitted_at":"2024-10-02T22:57:47Z","abstract_excerpt":"Posterior sampling in high-dimensional spaces using generative models holds significant promise for various applications, including but not limited to inverse problems and guided generation tasks. Despite many recent developments, generating diverse posterior samples remains a challenge, as existing methods require restarting the entire generative process for each new sample, making the procedure computationally expensive. In this work, we propose efficient posterior sampling by simulating Langevin dynamics in the noise space of a pre-trained generative model. By exploiting the mapping between"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.02078","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/2410.02078/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":"2410.02078","created_at":"2026-07-05T09:15:00.111052+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.02078v1","created_at":"2026-07-05T09:15:00.111052+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.02078","created_at":"2026-07-05T09:15:00.111052+00:00"},{"alias_kind":"pith_short_12","alias_value":"5JUNLTC42BDT","created_at":"2026-07-05T09:15:00.111052+00:00"},{"alias_kind":"pith_short_16","alias_value":"5JUNLTC42BDTJHVY","created_at":"2026-07-05T09:15:00.111052+00:00"},{"alias_kind":"pith_short_8","alias_value":"5JUNLTC4","created_at":"2026-07-05T09:15:00.111052+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2411.13443","citing_title":"Nonlinear Assimilation via Score-based Sequential Langevin Sampling","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2511.04859","citing_title":"Decoder-only Clustering in Attributed Graphs","ref_index":3,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5JUNLTC42BDTJHVYZZT5AWGIWA","json":"https://pith.science/pith/5JUNLTC42BDTJHVYZZT5AWGIWA.json","graph_json":"https://pith.science/api/pith-number/5JUNLTC42BDTJHVYZZT5AWGIWA/graph.json","events_json":"https://pith.science/api/pith-number/5JUNLTC42BDTJHVYZZT5AWGIWA/events.json","paper":"https://pith.science/paper/5JUNLTC4"},"agent_actions":{"view_html":"https://pith.science/pith/5JUNLTC42BDTJHVYZZT5AWGIWA","download_json":"https://pith.science/pith/5JUNLTC42BDTJHVYZZT5AWGIWA.json","view_paper":"https://pith.science/paper/5JUNLTC4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.02078&json=true","fetch_graph":"https://pith.science/api/pith-number/5JUNLTC42BDTJHVYZZT5AWGIWA/graph.json","fetch_events":"https://pith.science/api/pith-number/5JUNLTC42BDTJHVYZZT5AWGIWA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5JUNLTC42BDTJHVYZZT5AWGIWA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5JUNLTC42BDTJHVYZZT5AWGIWA/action/storage_attestation","attest_author":"https://pith.science/pith/5JUNLTC42BDTJHVYZZT5AWGIWA/action/author_attestation","sign_citation":"https://pith.science/pith/5JUNLTC42BDTJHVYZZT5AWGIWA/action/citation_signature","submit_replication":"https://pith.science/pith/5JUNLTC42BDTJHVYZZT5AWGIWA/action/replication_record"}},"created_at":"2026-07-05T09:15:00.111052+00:00","updated_at":"2026-07-05T09:15:00.111052+00:00"}