{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:ECDASA7ZPQCFJIQ7XFPYGBY3S5","short_pith_number":"pith:ECDASA7Z","schema_version":"1.0","canonical_sha256":"20860903f97c0454a21fb95f83071b97563d72163ea6c5ce233105d6a9e6c814","source":{"kind":"arxiv","id":"2607.04595","version":1},"attestation_state":"computed","paper":{"title":"Score Distributions, Not Cells: Evaluating Single-Cell Perturbations Under Class Overlap","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["q-bio.QM","stat.ML"],"primary_cat":"cs.LG","authors_text":"Davide D'Ascenzo, Sebastiano Cultrera di Montesano, Youssef Marrakchi","submitted_at":"2026-07-06T01:47:02Z","abstract_excerpt":"Most classification problems assume the classes are roughly separable, so that an individual sample can usually be assigned to one class. Single-cell perturbation data violates this assumption: two perturbations can produce different populations of cells while overlapping so much that an individual cell could belong to either. Per-cell accuracy then measures this overlap rather than model quality. We see this on Tahoe-100M and the Virtual Cell Challenge, where a linear classifier, an MLP, and a Transformer all plateau near macro-F1 0.2-0.3 even though almost every pair of perturbations is stat"},"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":"2607.04595","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-07-06T01:47:02Z","cross_cats_sorted":["q-bio.QM","stat.ML"],"title_canon_sha256":"40b1ad1a9255f33f17e973cc203669b72d01215b4df59de8715313e09893bf83","abstract_canon_sha256":"566017f9bd573dfc19545de98d216468346bf04dfaad66c6228e2018860452b1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-07T02:19:26.695401Z","signature_b64":"7zplmY+o9Xsy8N+QuhJyGh/GjgAefGgs4g/2K/3ZsuT82SyY6oiBCCg3Sos1HG+WgjrrU0VfNMUcNny39XK3DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"20860903f97c0454a21fb95f83071b97563d72163ea6c5ce233105d6a9e6c814","last_reissued_at":"2026-07-07T02:19:26.694784Z","signature_status":"signed_v1","first_computed_at":"2026-07-07T02:19:26.694784Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Score Distributions, Not Cells: Evaluating Single-Cell Perturbations Under Class Overlap","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["q-bio.QM","stat.ML"],"primary_cat":"cs.LG","authors_text":"Davide D'Ascenzo, Sebastiano Cultrera di Montesano, Youssef Marrakchi","submitted_at":"2026-07-06T01:47:02Z","abstract_excerpt":"Most classification problems assume the classes are roughly separable, so that an individual sample can usually be assigned to one class. Single-cell perturbation data violates this assumption: two perturbations can produce different populations of cells while overlapping so much that an individual cell could belong to either. Per-cell accuracy then measures this overlap rather than model quality. We see this on Tahoe-100M and the Virtual Cell Challenge, where a linear classifier, an MLP, and a Transformer all plateau near macro-F1 0.2-0.3 even though almost every pair of perturbations is stat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.04595","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/2607.04595/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":"2607.04595","created_at":"2026-07-07T02:19:26.694846+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.04595v1","created_at":"2026-07-07T02:19:26.694846+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.04595","created_at":"2026-07-07T02:19:26.694846+00:00"},{"alias_kind":"pith_short_12","alias_value":"ECDASA7ZPQCF","created_at":"2026-07-07T02:19:26.694846+00:00"},{"alias_kind":"pith_short_16","alias_value":"ECDASA7ZPQCFJIQ7","created_at":"2026-07-07T02:19:26.694846+00:00"},{"alias_kind":"pith_short_8","alias_value":"ECDASA7Z","created_at":"2026-07-07T02:19:26.694846+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/ECDASA7ZPQCFJIQ7XFPYGBY3S5","json":"https://pith.science/pith/ECDASA7ZPQCFJIQ7XFPYGBY3S5.json","graph_json":"https://pith.science/api/pith-number/ECDASA7ZPQCFJIQ7XFPYGBY3S5/graph.json","events_json":"https://pith.science/api/pith-number/ECDASA7ZPQCFJIQ7XFPYGBY3S5/events.json","paper":"https://pith.science/paper/ECDASA7Z"},"agent_actions":{"view_html":"https://pith.science/pith/ECDASA7ZPQCFJIQ7XFPYGBY3S5","download_json":"https://pith.science/pith/ECDASA7ZPQCFJIQ7XFPYGBY3S5.json","view_paper":"https://pith.science/paper/ECDASA7Z","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.04595&json=true","fetch_graph":"https://pith.science/api/pith-number/ECDASA7ZPQCFJIQ7XFPYGBY3S5/graph.json","fetch_events":"https://pith.science/api/pith-number/ECDASA7ZPQCFJIQ7XFPYGBY3S5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ECDASA7ZPQCFJIQ7XFPYGBY3S5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ECDASA7ZPQCFJIQ7XFPYGBY3S5/action/storage_attestation","attest_author":"https://pith.science/pith/ECDASA7ZPQCFJIQ7XFPYGBY3S5/action/author_attestation","sign_citation":"https://pith.science/pith/ECDASA7ZPQCFJIQ7XFPYGBY3S5/action/citation_signature","submit_replication":"https://pith.science/pith/ECDASA7ZPQCFJIQ7XFPYGBY3S5/action/replication_record"}},"created_at":"2026-07-07T02:19:26.694846+00:00","updated_at":"2026-07-07T02:19:26.694846+00:00"}