{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:ERJONTL6B2ZJXDIZQM33BH6RYQ","short_pith_number":"pith:ERJONTL6","schema_version":"1.0","canonical_sha256":"2452e6cd7e0eb29b8d198337b09fd1c4253a826c7e843715dd091d5766e0c003","source":{"kind":"arxiv","id":"2607.10508","version":1},"attestation_state":"computed","paper":{"title":"Confining Nondeterminism: AI-Driven Research Systems as DBMSs for Reliable, Non-Wasteful, Transparent, and Collaborative Research [Vision]","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.DB","authors_text":"Anastasia Ailamaki, Kyoungmin Kim","submitted_at":"2026-07-11T23:46:01Z","abstract_excerpt":"LLM agents that conduct research (proposing ideas, writing and running code, analyzing results) can already carry a study from research question to figures, yet cannot be fully trusted. The same question asked twice in a row returns different answers; the agent announces a number that no execution produced, and tool use does not prevent this, because nothing binds what the agent reports to what its tools returned; a small upstream change leaves downstream results silently stale, with no way to list which ones; and the agent re-runs preprocessing and rewrites code it has already produced. We ar"},"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":"2607.10508","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DB","submitted_at":"2026-07-11T23:46:01Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"1dc174e7bdb67cc8fed917f57420d21e7d00cdd3a1224dfd895d7eaf64490ec5","abstract_canon_sha256":"960e88cf8d3af9e20dbbb13e755bbe9b87602ff4e3d5d64f532157e34b0f0f80"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-14T01:21:18.851346Z","signature_b64":"8/OnS8ODnIcHO+Duote31HeLWBhxTFpYoMZAfcik94NI60QnwPDGorF7GRb27OKsjn+xeCFd8kkLFS43kPJeAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2452e6cd7e0eb29b8d198337b09fd1c4253a826c7e843715dd091d5766e0c003","last_reissued_at":"2026-07-14T01:21:18.850429Z","signature_status":"signed_v1","first_computed_at":"2026-07-14T01:21:18.850429Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Confining Nondeterminism: AI-Driven Research Systems as DBMSs for Reliable, Non-Wasteful, Transparent, and Collaborative Research [Vision]","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.DB","authors_text":"Anastasia Ailamaki, Kyoungmin Kim","submitted_at":"2026-07-11T23:46:01Z","abstract_excerpt":"LLM agents that conduct research (proposing ideas, writing and running code, analyzing results) can already carry a study from research question to figures, yet cannot be fully trusted. The same question asked twice in a row returns different answers; the agent announces a number that no execution produced, and tool use does not prevent this, because nothing binds what the agent reports to what its tools returned; a small upstream change leaves downstream results silently stale, with no way to list which ones; and the agent re-runs preprocessing and rewrites code it has already produced. We ar"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.10508","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/2607.10508/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":"2607.10508","created_at":"2026-07-14T01:21:18.850895+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.10508v1","created_at":"2026-07-14T01:21:18.850895+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.10508","created_at":"2026-07-14T01:21:18.850895+00:00"},{"alias_kind":"pith_short_12","alias_value":"ERJONTL6B2ZJ","created_at":"2026-07-14T01:21:18.850895+00:00"},{"alias_kind":"pith_short_16","alias_value":"ERJONTL6B2ZJXDIZ","created_at":"2026-07-14T01:21:18.850895+00:00"},{"alias_kind":"pith_short_8","alias_value":"ERJONTL6","created_at":"2026-07-14T01:21:18.850895+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/ERJONTL6B2ZJXDIZQM33BH6RYQ","json":"https://pith.science/pith/ERJONTL6B2ZJXDIZQM33BH6RYQ.json","graph_json":"https://pith.science/api/pith-number/ERJONTL6B2ZJXDIZQM33BH6RYQ/graph.json","events_json":"https://pith.science/api/pith-number/ERJONTL6B2ZJXDIZQM33BH6RYQ/events.json","paper":"https://pith.science/paper/ERJONTL6"},"agent_actions":{"view_html":"https://pith.science/pith/ERJONTL6B2ZJXDIZQM33BH6RYQ","download_json":"https://pith.science/pith/ERJONTL6B2ZJXDIZQM33BH6RYQ.json","view_paper":"https://pith.science/paper/ERJONTL6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.10508&json=true","fetch_graph":"https://pith.science/api/pith-number/ERJONTL6B2ZJXDIZQM33BH6RYQ/graph.json","fetch_events":"https://pith.science/api/pith-number/ERJONTL6B2ZJXDIZQM33BH6RYQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ERJONTL6B2ZJXDIZQM33BH6RYQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ERJONTL6B2ZJXDIZQM33BH6RYQ/action/storage_attestation","attest_author":"https://pith.science/pith/ERJONTL6B2ZJXDIZQM33BH6RYQ/action/author_attestation","sign_citation":"https://pith.science/pith/ERJONTL6B2ZJXDIZQM33BH6RYQ/action/citation_signature","submit_replication":"https://pith.science/pith/ERJONTL6B2ZJXDIZQM33BH6RYQ/action/replication_record"}},"created_at":"2026-07-14T01:21:18.850895+00:00","updated_at":"2026-07-14T01:21:18.850895+00:00"}