{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:2S3DN5VRH5TBK6VVTI4B3ZZECT","short_pith_number":"pith:2S3DN5VR","schema_version":"1.0","canonical_sha256":"d4b636f6b13f66157ab59a381de72414df08af5a587210f2097dd8ddafe3340c","source":{"kind":"arxiv","id":"2608.09696","version":1},"attestation_state":"computed","paper":{"title":"Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Kevin Murphy","submitted_at":"2026-08-10T14:59:22Z","abstract_excerpt":"Predicting the answer to interventional ``what if'' questions --- the outcome of an action never taken --- requires a \\emph{mechanistic}, causal model, not a curve fit; and learning such a model requires \\emph{experiments}, because passive data leaves its mechanisms unidentified. Experiments are expensive, so the central problem is \\emph{data efficiency}. We present the Model Discovery Agent (MDA), which couples a large language model (LLM), used as a \\emph{proposer} of candidate structures, with standard Bayesian machinery --- sequential Monte Carlo (SMC) for parameter and structure posterior"},"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":"2608.09696","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-10T14:59:22Z","cross_cats_sorted":[],"title_canon_sha256":"5d289b3bd6d966633760d1631d776deffedf7fccb0e47de135689effc2740a99","abstract_canon_sha256":"eb596f0546faab1bb42aad97439acc8dcf0a4d2027f47c3d6bf14d0d814799c4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-11T02:24:49.828508Z","signature_b64":"ZIqC4zVLFaN99mBW6K1YkwiX4WOobCP4taemydbJ3FrQc83MNEL72nPGbgOi5CqGO7HzKbekDDzvN9GekyCzBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d4b636f6b13f66157ab59a381de72414df08af5a587210f2097dd8ddafe3340c","last_reissued_at":"2026-08-11T02:24:49.824133Z","signature_status":"signed_v1","first_computed_at":"2026-08-11T02:24:49.824133Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Kevin Murphy","submitted_at":"2026-08-10T14:59:22Z","abstract_excerpt":"Predicting the answer to interventional ``what if'' questions --- the outcome of an action never taken --- requires a \\emph{mechanistic}, causal model, not a curve fit; and learning such a model requires \\emph{experiments}, because passive data leaves its mechanisms unidentified. Experiments are expensive, so the central problem is \\emph{data efficiency}. We present the Model Discovery Agent (MDA), which couples a large language model (LLM), used as a \\emph{proposer} of candidate structures, with standard Bayesian machinery --- sequential Monte Carlo (SMC) for parameter and structure posterior"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.09696","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/2608.09696/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":"2608.09696","created_at":"2026-08-11T02:24:49.827565+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.09696v1","created_at":"2026-08-11T02:24:49.827565+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.09696","created_at":"2026-08-11T02:24:49.827565+00:00"},{"alias_kind":"pith_short_12","alias_value":"2S3DN5VRH5TB","created_at":"2026-08-11T02:24:49.827565+00:00"},{"alias_kind":"pith_short_16","alias_value":"2S3DN5VRH5TBK6VV","created_at":"2026-08-11T02:24:49.827565+00:00"},{"alias_kind":"pith_short_8","alias_value":"2S3DN5VR","created_at":"2026-08-11T02:24:49.827565+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2S3DN5VRH5TBK6VVTI4B3ZZECT","json":"https://pith.science/pith/2S3DN5VRH5TBK6VVTI4B3ZZECT.json","graph_json":"https://pith.science/api/pith-number/2S3DN5VRH5TBK6VVTI4B3ZZECT/graph.json","events_json":"https://pith.science/api/pith-number/2S3DN5VRH5TBK6VVTI4B3ZZECT/events.json","paper":"https://pith.science/paper/2S3DN5VR"},"agent_actions":{"view_html":"https://pith.science/pith/2S3DN5VRH5TBK6VVTI4B3ZZECT","download_json":"https://pith.science/pith/2S3DN5VRH5TBK6VVTI4B3ZZECT.json","view_paper":"https://pith.science/paper/2S3DN5VR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.09696&json=true","fetch_graph":"https://pith.science/api/pith-number/2S3DN5VRH5TBK6VVTI4B3ZZECT/graph.json","fetch_events":"https://pith.science/api/pith-number/2S3DN5VRH5TBK6VVTI4B3ZZECT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2S3DN5VRH5TBK6VVTI4B3ZZECT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2S3DN5VRH5TBK6VVTI4B3ZZECT/action/storage_attestation","attest_author":"https://pith.science/pith/2S3DN5VRH5TBK6VVTI4B3ZZECT/action/author_attestation","sign_citation":"https://pith.science/pith/2S3DN5VRH5TBK6VVTI4B3ZZECT/action/citation_signature","submit_replication":"https://pith.science/pith/2S3DN5VRH5TBK6VVTI4B3ZZECT/action/replication_record"}},"created_at":"2026-08-11T02:24:49.827565+00:00","updated_at":"2026-08-11T02:24:49.827565+00:00"}