{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:AJBXFCBCSXDQH4M35C4QRWTXEJ","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":"fd67c6fce58c49e9040aef8478e7ccbcc5cf48010ddc39298caaa3a75da8be01","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-17T22:05:38Z","title_canon_sha256":"f3cdb279340a1f0f9d93c64b112db494dfb97897de45a5b643888dd3673d378b"},"schema_version":"1.0","source":{"id":"2606.19624","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.19624","created_at":"2026-06-19T16:12:30Z"},{"alias_kind":"arxiv_version","alias_value":"2606.19624v1","created_at":"2026-06-19T16:12:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.19624","created_at":"2026-06-19T16:12:30Z"},{"alias_kind":"pith_short_12","alias_value":"AJBXFCBCSXDQ","created_at":"2026-06-19T16:12:30Z"},{"alias_kind":"pith_short_16","alias_value":"AJBXFCBCSXDQH4M3","created_at":"2026-06-19T16:12:30Z"},{"alias_kind":"pith_short_8","alias_value":"AJBXFCBC","created_at":"2026-06-19T16:12:30Z"}],"graph_snapshots":[{"event_id":"sha256:195af1115230d890f4d87e0e68577aa699e61ccfd1dbe89e1f1acac58a1dd64b","target":"graph","created_at":"2026-06-19T16:12:30Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2606.19624/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reliable benchmarking is critical for developing machine learning models for tandem mass spectrometry (MS/MS) based molecule discovery. Subtle issues in experimental design and model evaluation procedures can degrade the trustworthiness of such benchmarks and lead to erroneous conclusions. We conduct a thorough review of model evaluation issues in the recent MS/MS machine learning literature, using the standard MassSpecGym benchmark suite as a case study to illustrate the impact of these issues. We find evaluation issues in at least 17 of 26 papers reporting MassSpecGym benchmark results in th","authors_text":"Anton Bushuiev, Connor W. Coley, David Healey, Filip Jozefov, Hongxuan Liu, Ivy Lightheart, James Taylor, Josef Sivic, Magdalena Lederbauer, Mrunali Manjrekar, Roman Bushuiev, Runzhong Wang, Soha Hassoun, Tom\\'a\\v{s} Pluskal, Yinkai Wang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-17T22:05:38Z","title":"MassSpecGym in the Wild: Uncovering and Correcting Evaluation Pitfalls in AI-Driven Molecule Discovery"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.19624","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:8072b757e8b266ee88288d634562ab36ac007c2ac4c6396dd7537fde0689ca3c","target":"record","created_at":"2026-06-19T16:12:30Z","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":"fd67c6fce58c49e9040aef8478e7ccbcc5cf48010ddc39298caaa3a75da8be01","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-17T22:05:38Z","title_canon_sha256":"f3cdb279340a1f0f9d93c64b112db494dfb97897de45a5b643888dd3673d378b"},"schema_version":"1.0","source":{"id":"2606.19624","kind":"arxiv","version":1}},"canonical_sha256":"024372882295c703f19be8b908da772269ca9d35de3862d1da4f814292243556","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"024372882295c703f19be8b908da772269ca9d35de3862d1da4f814292243556","first_computed_at":"2026-06-19T16:12:30.549352Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-19T16:12:30.549352Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"yLo3TDCzfg84CVfFkkFJuB4fIdafq1ZC7tXmo6kJzzDsdSOoCdYrTzbN256xUPUEfkHAlOGl5Msi9L4c2HQ8AA==","signature_status":"signed_v1","signed_at":"2026-06-19T16:12:30.549758Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.19624","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8072b757e8b266ee88288d634562ab36ac007c2ac4c6396dd7537fde0689ca3c","sha256:195af1115230d890f4d87e0e68577aa699e61ccfd1dbe89e1f1acac58a1dd64b"],"state_sha256":"9fd70a2d28302bfe1109d675b29535608d03a4af8728758e3a97673c56e7e24b"}