{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:PTZFZGGGL32EJMC2AR5SVGM76F","short_pith_number":"pith:PTZFZGGG","schema_version":"1.0","canonical_sha256":"7cf25c98c65ef444b05a047b2a999ff16bab1f7f4c95f3b9e58cf309502498ab","source":{"kind":"arxiv","id":"2110.06581","version":3},"attestation_state":"computed","paper":{"title":"A Trust Crisis In Simulation-Based Inference? Your Posterior Approximations Can Be Unfaithful","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Antoine Wehenkel, Arnaud Delaunoy, Fran\\c{c}ois Rozet, Gilles Louppe, Joeri Hermans, Volodimir Begy","submitted_at":"2021-10-13T08:50:05Z","abstract_excerpt":"We present extensive empirical evidence showing that current Bayesian simulation-based inference algorithms can produce computationally unfaithful posterior approximations. Our results show that all benchmarked algorithms -- (Sequential) Neural Posterior Estimation, (Sequential) Neural Ratio Estimation, Sequential Neural Likelihood and variants of Approximate Bayesian Computation -- can yield overconfident posterior approximations, which makes them unreliable for scientific use cases and falsificationist inquiry. Failing to address this issue may reduce the range of applicability of simulation"},"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":"2110.06581","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2021-10-13T08:50:05Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"9ecbc04804447627e77d021e809645001a2ac77ce04abce3c6548cee3b56a3a0","abstract_canon_sha256":"9e58dea5727c43579f47399d1ef93de5790e5f1696ca9520133a9292eec7b772"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:22:04.744769Z","signature_b64":"zMD+eIQSKRjAkEWloP21aOiZuXbXfyFTXPvDHWItUczzijU1nfEUDdpzUQNURRjq3595oYf1cFMd+Eu8/d+gAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7cf25c98c65ef444b05a047b2a999ff16bab1f7f4c95f3b9e58cf309502498ab","last_reissued_at":"2026-07-05T05:22:04.744299Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:22:04.744299Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Trust Crisis In Simulation-Based Inference? Your Posterior Approximations Can Be Unfaithful","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Antoine Wehenkel, Arnaud Delaunoy, Fran\\c{c}ois Rozet, Gilles Louppe, Joeri Hermans, Volodimir Begy","submitted_at":"2021-10-13T08:50:05Z","abstract_excerpt":"We present extensive empirical evidence showing that current Bayesian simulation-based inference algorithms can produce computationally unfaithful posterior approximations. Our results show that all benchmarked algorithms -- (Sequential) Neural Posterior Estimation, (Sequential) Neural Ratio Estimation, Sequential Neural Likelihood and variants of Approximate Bayesian Computation -- can yield overconfident posterior approximations, which makes them unreliable for scientific use cases and falsificationist inquiry. Failing to address this issue may reduce the range of applicability of simulation"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.06581","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/2110.06581/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":"2110.06581","created_at":"2026-07-05T05:22:04.744371+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.06581v3","created_at":"2026-07-05T05:22:04.744371+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.06581","created_at":"2026-07-05T05:22:04.744371+00:00"},{"alias_kind":"pith_short_12","alias_value":"PTZFZGGGL32E","created_at":"2026-07-05T05:22:04.744371+00:00"},{"alias_kind":"pith_short_16","alias_value":"PTZFZGGGL32EJMC2","created_at":"2026-07-05T05:22:04.744371+00:00"},{"alias_kind":"pith_short_8","alias_value":"PTZFZGGG","created_at":"2026-07-05T05:22:04.744371+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":9,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.25446","citing_title":"Amortized Simulation-Based Inference of Relativistic Mean-Field Couplings for Neutron-Star Equations of State","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2606.11309","citing_title":"Dark Energy Survey Year 3 results: optimized $w$CDM simulation-based inference with weak lensing map-level hybrid statistics","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2606.12255","citing_title":"Towards Practical Field-Level Inference for Weak Lensing","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22891","citing_title":"Pointwise Metrics Mislead: An Evaluation Protocol for Multimodal Inverse Problems","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2410.00755","citing_title":"Model-independent searches of new physics in DARWIN with a semi-supervised deep learning pipeline","ref_index":90,"is_internal_anchor":false},{"citing_arxiv_id":"2506.09374","citing_title":"Inherited or produced? Inferring protein production kinetics when protein counts are shaped by a cell's division history","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2512.23748","citing_title":"A Review of Diffusion-based Simulation-Based Inference: Foundations and Applications in Non-Ideal Data Scenarios","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06790","citing_title":"Machine Learning Techniques for Astrophysics and Cosmology: Photometric Redshifts","ref_index":172,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19919","citing_title":"Neural Simulation-based Inference with Hierarchical Priors for Detached Eclipsing Binaries","ref_index":11,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PTZFZGGGL32EJMC2AR5SVGM76F","json":"https://pith.science/pith/PTZFZGGGL32EJMC2AR5SVGM76F.json","graph_json":"https://pith.science/api/pith-number/PTZFZGGGL32EJMC2AR5SVGM76F/graph.json","events_json":"https://pith.science/api/pith-number/PTZFZGGGL32EJMC2AR5SVGM76F/events.json","paper":"https://pith.science/paper/PTZFZGGG"},"agent_actions":{"view_html":"https://pith.science/pith/PTZFZGGGL32EJMC2AR5SVGM76F","download_json":"https://pith.science/pith/PTZFZGGGL32EJMC2AR5SVGM76F.json","view_paper":"https://pith.science/paper/PTZFZGGG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.06581&json=true","fetch_graph":"https://pith.science/api/pith-number/PTZFZGGGL32EJMC2AR5SVGM76F/graph.json","fetch_events":"https://pith.science/api/pith-number/PTZFZGGGL32EJMC2AR5SVGM76F/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PTZFZGGGL32EJMC2AR5SVGM76F/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PTZFZGGGL32EJMC2AR5SVGM76F/action/storage_attestation","attest_author":"https://pith.science/pith/PTZFZGGGL32EJMC2AR5SVGM76F/action/author_attestation","sign_citation":"https://pith.science/pith/PTZFZGGGL32EJMC2AR5SVGM76F/action/citation_signature","submit_replication":"https://pith.science/pith/PTZFZGGGL32EJMC2AR5SVGM76F/action/replication_record"}},"created_at":"2026-07-05T05:22:04.744371+00:00","updated_at":"2026-07-05T05:22:04.744371+00:00"}