{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:24LU7BDEXJAHPZ7YK5SPGXYSK4","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"e5523be2341fc8483708e9c567e797f5b7198cab93fc02306b5706338ce82398","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-13T09:30:56Z","title_canon_sha256":"45d9594ca99e8d5f4a3d67c360e4a42fb169774ce394c842130f0680269ae33b"},"schema_version":"1.0","source":{"id":"2605.13245","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.13245","created_at":"2026-05-18T02:44:49Z"},{"alias_kind":"arxiv_version","alias_value":"2605.13245v1","created_at":"2026-05-18T02:44:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.13245","created_at":"2026-05-18T02:44:49Z"},{"alias_kind":"pith_short_12","alias_value":"24LU7BDEXJAH","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"24LU7BDEXJAHPZ7Y","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"24LU7BDE","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:6a72d050ca0c9b5229732e431bb1a7f863d091a94eb178baccca453b3994631f","target":"graph","created_at":"2026-05-18T02:44:49Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"The typed tool produces identical results across all runs. The commercial platforms either vary in numerical output and analytical methodology across runs, or fail to produce valid results on the task."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That a researcher's exact analytical procedure for a given instrument can be fully captured as a deterministic tool through structured interviews, without loss of nuance or need for ongoing human judgment."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Typed mediation lets LLMs orchestrate fixed deterministic tools for scientific workflows, producing identical analysis outputs across runs where direct LLM generation varies."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Language models achieve identical scientific analysis results every time by calling fixed deterministic tools rather than generating code."}],"snapshot_sha256":"ee668654f6c49df4bbf1e4639ad74195fc908e01806e3b1af3cd3d40295d6926"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"Language models can produce convincing scientific analyses, but repeated generations on the same data do not guarantee the same result. A researcher may regenerate an identical query and receive a different fit, a different peak position or a different analysis procedure, without an obvious way to decide which output to trust. We propose typed mediation, a pattern in which the model orchestrates deterministic tools rather than generating analytical code. Each tool encodes one researcher's exact procedure for one instrument, ported through structured interviews. The model selects which tool to ","authors_text":"Danae Katrisioti, Emmanuel Stratakis, Marios Adamidis, Yannis Tzitzikas","cross_cats":[],"headline":"Language models achieve identical scientific analysis results every time by calling fixed deterministic tools rather than generating code.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-13T09:30:56Z","title":"It's not the Language Model, it's the Tool: Deterministic Mediation for Scientific Workflows"},"references":{"count":24,"internal_anchors":0,"resolved_work":24,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"Defeating nondeterminism in LLM inference, September 2025","work_id":"2c3260f3-471d-4312-8ad3-90156eda0adb","year":2025},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Measuring AI agent autonomy in practice, February 2026","work_id":"572e02da-4eec-46d4-b95f-6bcc839592c9","year":2026},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"John R. Kitchin. The evolving role of programming and LLMs in the development of self-driving laboratories.APL Machine Learning, 3(2):026111, 2025","work_id":"a79f8f1f-903b-4ce0-ad1b-25c1f6dc6258","year":2025},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Leak, cheat, repeat: Data contamination and evaluation malpractices in closed-source LLMs","work_id":"488f0670-6a95-4dc9-8f66-886650e2c48f","year":2024},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Leemann, Antonin Sulc, and Marco Venturini","work_id":"21c00416-9803-44f8-a449-1d5b28f75020","year":2025}],"snapshot_sha256":"aade5c7f48c45a03e59b43db503e1c6dd11db10268d63b038d31417bc1b239ce"},"source":{"id":"2605.13245","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-14T19:52:31.635103Z","id":"4acfe267-bcd7-4679-92b8-35a10b9f7acd","model_set":{"reader":"grok-4.3"},"one_line_summary":"Typed mediation lets LLMs orchestrate fixed deterministic tools for scientific workflows, producing identical analysis outputs across runs where direct LLM generation varies.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Language models achieve identical scientific analysis results every time by calling fixed deterministic tools rather than generating code.","strongest_claim":"The typed tool produces identical results across all runs. The commercial platforms either vary in numerical output and analytical methodology across runs, or fail to produce valid results on the task.","weakest_assumption":"That a researcher's exact analytical procedure for a given instrument can be fully captured as a deterministic tool through structured interviews, without loss of nuance or need for ongoing human judgment."}},"verdict_id":"4acfe267-bcd7-4679-92b8-35a10b9f7acd"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:356315d25e5b33ff6b7bfe5f5b1075b854c7e7263bbff84b7a736e53314dd5b5","target":"record","created_at":"2026-05-18T02:44:49Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"e5523be2341fc8483708e9c567e797f5b7198cab93fc02306b5706338ce82398","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-13T09:30:56Z","title_canon_sha256":"45d9594ca99e8d5f4a3d67c360e4a42fb169774ce394c842130f0680269ae33b"},"schema_version":"1.0","source":{"id":"2605.13245","kind":"arxiv","version":1}},"canonical_sha256":"d7174f8464ba4077e7f85764f35f12572b49440ef8c8bf59e2a9593d8d625ac7","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d7174f8464ba4077e7f85764f35f12572b49440ef8c8bf59e2a9593d8d625ac7","first_computed_at":"2026-05-18T02:44:49.471948Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T02:44:49.471948Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"0mzJWvE6XNt3mgmJY3x4yDuy+SgbMcer3b8ax3+iIVDKVNWOSXO4hHxXOwvhzd5at3n5cjyXALW6zYRIUSl7BQ==","signature_status":"signed_v1","signed_at":"2026-05-18T02:44:49.472397Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.13245","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:356315d25e5b33ff6b7bfe5f5b1075b854c7e7263bbff84b7a736e53314dd5b5","sha256:6a72d050ca0c9b5229732e431bb1a7f863d091a94eb178baccca453b3994631f"],"state_sha256":"e59d2f7c10e5239addc7b28d7a52282245d18016df5c260aaf96ab88c34c9648"}