{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:CA7G57KQJ3KZCRQKMC2IKDPJ4Z","short_pith_number":"pith:CA7G57KQ","schema_version":"1.0","canonical_sha256":"103e6efd504ed591460a60b4850de9e67fccb5e8d543950c032a020d50a24213","source":{"kind":"arxiv","id":"2502.15224","version":2},"attestation_state":"computed","paper":{"title":"Auto-Discovery-Bench: Diagnosing Structured State Tracking in Oracle-Guided Discovery","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Anirudh Goyal, Beibei Lin, Dianbo Liu, Qiran Zou, Srinivas Anumasa, Tingting Chen, Vedant Shah, Zifeng Yuan","submitted_at":"2025-02-21T05:35:20Z","abstract_excerpt":"Interactive discovery requires agents to maintain and update structured beliefs over many rounds of feedback. Before evaluating agents in noisy, open-ended scientific environments, it is useful to isolate this prerequisite capability under controlled conditions. We introduce Auto-Discovery-Bench, a deterministic oracle-guided diagnostic benchmark in which agents recover hidden structures through repeated hypothesis--intervention--feedback cycles. The benchmark instantiates three controlled discovery abstractions: directed graph discovery, undirected relational discovery, and symbolic equation "},"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":"2502.15224","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-21T05:35:20Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"cf92ca72477970bb06061a179b8bd8805b85e946faf1143b3d1fd2be39f9a34b","abstract_canon_sha256":"f95b0bf059b78ecc1b2e59d9e6aa38d0a00d7cc9ff6a8b73f978a2984182829f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-01T01:03:39.695399Z","signature_b64":"tj6kkuSZSunY5sWDRXQSOn2/8vVyfUAkEauqBozi1fhu03Pq/IWuhlMe2d657sBDgqSGjz+NRqfopUlqvSviDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"103e6efd504ed591460a60b4850de9e67fccb5e8d543950c032a020d50a24213","last_reissued_at":"2026-06-01T01:03:39.694676Z","signature_status":"signed_v1","first_computed_at":"2026-06-01T01:03:39.694676Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Auto-Discovery-Bench: Diagnosing Structured State Tracking in Oracle-Guided Discovery","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Anirudh Goyal, Beibei Lin, Dianbo Liu, Qiran Zou, Srinivas Anumasa, Tingting Chen, Vedant Shah, Zifeng Yuan","submitted_at":"2025-02-21T05:35:20Z","abstract_excerpt":"Interactive discovery requires agents to maintain and update structured beliefs over many rounds of feedback. Before evaluating agents in noisy, open-ended scientific environments, it is useful to isolate this prerequisite capability under controlled conditions. We introduce Auto-Discovery-Bench, a deterministic oracle-guided diagnostic benchmark in which agents recover hidden structures through repeated hypothesis--intervention--feedback cycles. The benchmark instantiates three controlled discovery abstractions: directed graph discovery, undirected relational discovery, and symbolic equation "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.15224","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/2502.15224/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":"2502.15224","created_at":"2026-06-01T01:03:39.694774+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.15224v2","created_at":"2026-06-01T01:03:39.694774+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.15224","created_at":"2026-06-01T01:03:39.694774+00:00"},{"alias_kind":"pith_short_12","alias_value":"CA7G57KQJ3KZ","created_at":"2026-06-01T01:03:39.694774+00:00"},{"alias_kind":"pith_short_16","alias_value":"CA7G57KQJ3KZCRQK","created_at":"2026-06-01T01:03:39.694774+00:00"},{"alias_kind":"pith_short_8","alias_value":"CA7G57KQ","created_at":"2026-06-01T01:03:39.694774+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2605.23602","citing_title":"GlowGS: Generative Semantic Feature Learning for 3D Gaussian Splatting in Nighttime Glow Scenes","ref_index":6,"is_internal_anchor":true},{"citing_arxiv_id":"2605.23204","citing_title":"AutoResearch AI: Towards AI-Powered Research Automation for Scientific Discovery","ref_index":54,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CA7G57KQJ3KZCRQKMC2IKDPJ4Z","json":"https://pith.science/pith/CA7G57KQJ3KZCRQKMC2IKDPJ4Z.json","graph_json":"https://pith.science/api/pith-number/CA7G57KQJ3KZCRQKMC2IKDPJ4Z/graph.json","events_json":"https://pith.science/api/pith-number/CA7G57KQJ3KZCRQKMC2IKDPJ4Z/events.json","paper":"https://pith.science/paper/CA7G57KQ"},"agent_actions":{"view_html":"https://pith.science/pith/CA7G57KQJ3KZCRQKMC2IKDPJ4Z","download_json":"https://pith.science/pith/CA7G57KQJ3KZCRQKMC2IKDPJ4Z.json","view_paper":"https://pith.science/paper/CA7G57KQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.15224&json=true","fetch_graph":"https://pith.science/api/pith-number/CA7G57KQJ3KZCRQKMC2IKDPJ4Z/graph.json","fetch_events":"https://pith.science/api/pith-number/CA7G57KQJ3KZCRQKMC2IKDPJ4Z/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CA7G57KQJ3KZCRQKMC2IKDPJ4Z/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CA7G57KQJ3KZCRQKMC2IKDPJ4Z/action/storage_attestation","attest_author":"https://pith.science/pith/CA7G57KQJ3KZCRQKMC2IKDPJ4Z/action/author_attestation","sign_citation":"https://pith.science/pith/CA7G57KQJ3KZCRQKMC2IKDPJ4Z/action/citation_signature","submit_replication":"https://pith.science/pith/CA7G57KQJ3KZCRQKMC2IKDPJ4Z/action/replication_record"}},"created_at":"2026-06-01T01:03:39.694774+00:00","updated_at":"2026-06-01T01:03:39.694774+00:00"}