{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:YJHUBZ4A53NWZCTORMSRZR7CDT","short_pith_number":"pith:YJHUBZ4A","schema_version":"1.0","canonical_sha256":"c24f40e780eedb6c8a6e8b251cc7e21ccfb187c2fcab4fb1c70f1b2824ddd4c4","source":{"kind":"arxiv","id":"2512.22999","version":2},"attestation_state":"computed","paper":{"title":"JADAI: Jointly Amortizing Adaptive Design and Bayesian Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"stat.ML","authors_text":"Desi R. Ivanova, Lars K\\\"uhmichel, Niels Bracher, Paul-Christian B\\\"urkner, Stefan T. Radev, Xavier Intes","submitted_at":"2025-12-28T16:54:43Z","abstract_excerpt":"We consider problems of parameter estimation where design variables can be actively optimized to maximize information gain. To this end, we introduce JADAI, a framework that jointly amortizes Bayesian adaptive design and inference by training a policy, a history network, and an inference network end-to-end. The networks minimize a generic loss that aggregates incremental reductions in posterior error along experimental sequences. Inference networks are instantiated with diffusion-based posterior estimators that can approximate high-dimensional and multimodal posteriors at every experimental st"},"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":"2512.22999","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2025-12-28T16:54:43Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"063b505d789bce519987927389d506141c4951298a106c9ce6bfb28498061b36","abstract_canon_sha256":"c41fbb13b2eb1b4b7582988e7a2693ca9593996ea04842fd60eae71819d0bc1c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-07T00:15:51.682932Z","signature_b64":"pM6IQpC5dWGqQ35RaHI1ENjphvdfh7wzJzxG22HSLZAaSx4sggaOyTrdBnApYggn618OIW4Tqo+2XNL6kZWFCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c24f40e780eedb6c8a6e8b251cc7e21ccfb187c2fcab4fb1c70f1b2824ddd4c4","last_reissued_at":"2026-07-07T00:15:51.681998Z","signature_status":"signed_v1","first_computed_at":"2026-07-07T00:15:51.681998Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"JADAI: Jointly Amortizing Adaptive Design and Bayesian Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"stat.ML","authors_text":"Desi R. Ivanova, Lars K\\\"uhmichel, Niels Bracher, Paul-Christian B\\\"urkner, Stefan T. Radev, Xavier Intes","submitted_at":"2025-12-28T16:54:43Z","abstract_excerpt":"We consider problems of parameter estimation where design variables can be actively optimized to maximize information gain. To this end, we introduce JADAI, a framework that jointly amortizes Bayesian adaptive design and inference by training a policy, a history network, and an inference network end-to-end. The networks minimize a generic loss that aggregates incremental reductions in posterior error along experimental sequences. Inference networks are instantiated with diffusion-based posterior estimators that can approximate high-dimensional and multimodal posteriors at every experimental st"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2512.22999","kind":"arxiv","version":2},"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/2512.22999/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":"2512.22999","created_at":"2026-07-07T00:15:51.682140+00:00"},{"alias_kind":"arxiv_version","alias_value":"2512.22999v2","created_at":"2026-07-07T00:15:51.682140+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2512.22999","created_at":"2026-07-07T00:15:51.682140+00:00"},{"alias_kind":"pith_short_12","alias_value":"YJHUBZ4A53NW","created_at":"2026-07-07T00:15:51.682140+00:00"},{"alias_kind":"pith_short_16","alias_value":"YJHUBZ4A53NWZCTO","created_at":"2026-07-07T00:15:51.682140+00:00"},{"alias_kind":"pith_short_8","alias_value":"YJHUBZ4A","created_at":"2026-07-07T00:15:51.682140+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":6,"sample":[{"citing_arxiv_id":"2607.08335","citing_title":"Bayesian Experimental Design via Score Matching","ref_index":194,"is_internal_anchor":true},{"citing_arxiv_id":"2607.08335","citing_title":"Bayesian Experimental Design via Score Matching","ref_index":9,"is_internal_anchor":true},{"citing_arxiv_id":"2606.25197","citing_title":"Efficient Adaptive Data Acquisition via Pretrained Belief Representations","ref_index":7,"is_internal_anchor":true},{"citing_arxiv_id":"2606.23662","citing_title":"Action-BED: Task-Driven Bayesian Experimental Design with Singly Intractable Objectives","ref_index":10,"is_internal_anchor":true},{"citing_arxiv_id":"2606.23515","citing_title":"FairBED: A Bayesian Experimental Design Approach to Gathering Fairer Data","ref_index":13,"is_internal_anchor":true},{"citing_arxiv_id":"2604.18319","citing_title":"Overcoming Selection Bias in Statistical Studies With Amortized Bayesian Inference","ref_index":66,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YJHUBZ4A53NWZCTORMSRZR7CDT","json":"https://pith.science/pith/YJHUBZ4A53NWZCTORMSRZR7CDT.json","graph_json":"https://pith.science/api/pith-number/YJHUBZ4A53NWZCTORMSRZR7CDT/graph.json","events_json":"https://pith.science/api/pith-number/YJHUBZ4A53NWZCTORMSRZR7CDT/events.json","paper":"https://pith.science/paper/YJHUBZ4A"},"agent_actions":{"view_html":"https://pith.science/pith/YJHUBZ4A53NWZCTORMSRZR7CDT","download_json":"https://pith.science/pith/YJHUBZ4A53NWZCTORMSRZR7CDT.json","view_paper":"https://pith.science/paper/YJHUBZ4A","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2512.22999&json=true","fetch_graph":"https://pith.science/api/pith-number/YJHUBZ4A53NWZCTORMSRZR7CDT/graph.json","fetch_events":"https://pith.science/api/pith-number/YJHUBZ4A53NWZCTORMSRZR7CDT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YJHUBZ4A53NWZCTORMSRZR7CDT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YJHUBZ4A53NWZCTORMSRZR7CDT/action/storage_attestation","attest_author":"https://pith.science/pith/YJHUBZ4A53NWZCTORMSRZR7CDT/action/author_attestation","sign_citation":"https://pith.science/pith/YJHUBZ4A53NWZCTORMSRZR7CDT/action/citation_signature","submit_replication":"https://pith.science/pith/YJHUBZ4A53NWZCTORMSRZR7CDT/action/replication_record"}},"created_at":"2026-07-07T00:15:51.682140+00:00","updated_at":"2026-07-07T00:15:51.682140+00:00"}