{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:PCECLMH6ZFNBLTQ7HY77MOKVDB","short_pith_number":"pith:PCECLMH6","schema_version":"1.0","canonical_sha256":"788825b0fec95a15ce1f3e3ff63955184d1a166a4416f57cac518364f77926fb","source":{"kind":"arxiv","id":"2605.16104","version":1},"attestation_state":"computed","paper":{"title":"StateXDiff: Cell State-Contextualized Multimodal Diffusion for Single-Cell Perturbation Prediction","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"StateXDiff predicts single-cell drug responses more accurately under out-of-distribution conditions by integrating transcriptomic and inferred protein features through conditional diffusion.","cross_cats":["q-bio.QM"],"primary_cat":"q-bio.GN","authors_text":"Jianzhong Jeff Xi, Ningfeng Que, Peiting Shi, Xianzhe Huang, Xiaofei Wang","submitted_at":"2026-05-15T15:54:46Z","abstract_excerpt":"Predicting drug-induced cellular state changes at single-cell resolution remains a central challenge in virtual cell modeling, particularly under out-of-distribution (OOD) conditions. Current approaches predominantly rely on RNA-based assays, which often fail to adequately capture the diverse cellular states underlying drug responses. Moreover, conditional distribution shifts and low signal-to-noise ratios frequently cause models to learn spurious correlations rather than genuine state transitions. To address these limitations, we introduce StateXDiff, a cell State-contextualized multimodal (X"},"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":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2605.16104","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"q-bio.GN","submitted_at":"2026-05-15T15:54:46Z","cross_cats_sorted":["q-bio.QM"],"title_canon_sha256":"b1747cdae1c3e8838010daa4a21d857ca5bfbfed704b16c22573394f37207429","abstract_canon_sha256":"0ccc779287fa8f5f6b3bd9e1a21eae1bf189885f50645e21e3e912e6c0cb2c90"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:01:52.845095Z","signature_b64":"utusd7wmcGh1RxD/evPWM4Fnmo6mw3McODp2feIGynjP/bQ3lnebrNeKeGjG3fjn9X1mi55GOoCTGdLxI+CiDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"788825b0fec95a15ce1f3e3ff63955184d1a166a4416f57cac518364f77926fb","last_reissued_at":"2026-05-20T00:01:52.844408Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:01:52.844408Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"StateXDiff: Cell State-Contextualized Multimodal Diffusion for Single-Cell Perturbation Prediction","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"StateXDiff predicts single-cell drug responses more accurately under out-of-distribution conditions by integrating transcriptomic and inferred protein features through conditional diffusion.","cross_cats":["q-bio.QM"],"primary_cat":"q-bio.GN","authors_text":"Jianzhong Jeff Xi, Ningfeng Que, Peiting Shi, Xianzhe Huang, Xiaofei Wang","submitted_at":"2026-05-15T15:54:46Z","abstract_excerpt":"Predicting drug-induced cellular state changes at single-cell resolution remains a central challenge in virtual cell modeling, particularly under out-of-distribution (OOD) conditions. Current approaches predominantly rely on RNA-based assays, which often fail to adequately capture the diverse cellular states underlying drug responses. Moreover, conditional distribution shifts and low signal-to-noise ratios frequently cause models to learn spurious correlations rather than genuine state transitions. To address these limitations, we introduce StateXDiff, a cell State-contextualized multimodal (X"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"StateXDiff consistently enhances generalization performance across three challenging settings: unseen cell lines, unseen drugs, and combinatorial perturbations.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the inferred protein features, when combined with transcriptomic profiles, produce a disentangled representation that captures genuine biological state transitions rather than spurious correlations induced by conditional distribution shifts or low signal-to-noise ratios.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"StateXDiff integrates transcriptomic profiles with inferred protein features via a conditional diffusion model and mechanism-aware drug templates to predict single-cell drug perturbation responses under unseen cell lines, drugs, and combinatorial settings.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"StateXDiff predicts single-cell drug responses more accurately under out-of-distribution conditions by integrating transcriptomic and inferred protein features through conditional diffusion.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"a90f9a90d1c8d045c33199f72a878a14cd22f1140a91b9f21903b419e2ca5461"},"source":{"id":"2605.16104","kind":"arxiv","version":1},"verdict":{"id":"6ff07fae-2c28-4117-8a28-a00781fb2c2e","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T17:03:21.413491Z","strongest_claim":"StateXDiff consistently enhances generalization performance across three challenging settings: unseen cell lines, unseen drugs, and combinatorial perturbations.","one_line_summary":"StateXDiff integrates transcriptomic profiles with inferred protein features via a conditional diffusion model and mechanism-aware drug templates to predict single-cell drug perturbation responses under unseen cell lines, drugs, and combinatorial settings.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the inferred protein features, when combined with transcriptomic profiles, produce a disentangled representation that captures genuine biological state transitions rather than spurious correlations induced by conditional distribution shifts or low signal-to-noise ratios.","pith_extraction_headline":"StateXDiff predicts single-cell drug responses more accurately under out-of-distribution conditions by integrating transcriptomic and inferred protein features through conditional diffusion."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.16104/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-19T17:33:38.070410Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T17:31:18.417669Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T17:16:19.341090Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T16:41:55.486272Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"add6a19ee620495a9c62358f4a6a4346205ec16b8f0dfa4f9cf302ecb633f164"},"references":{"count":28,"sample":[{"doi":"","year":2024,"title":"How to build the virtual cell with artificial intelligence: Priorities and opportunities.Cell, 187(25):7045–7063, 2024","work_id":"3730d097-4e49-4e24-90b2-08d3e80512d9","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"Digital twins in oncology: where we are and where we hope to go.BMJ oncology, 4(1):e000893, 2025","work_id":"cf9c965b-bd49-446a-9c1f-f7da00216774","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"Deep-learning-based gene perturbation effect prediction does not yet outperform simple linear baselines.Nature Methods, 22(8):1657–1661, 2025","work_id":"8fb3a187-dbfd-478b-a071-99727993fd58","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2017,"title":"A next generation connectivity map: L1000 platform and the first 1,000,000 profiles","work_id":"a056cecd-a0f5-46a0-bb78-b3d5235dd59e","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"A pre-trained large generative model for translating single-cell transcriptomes to proteomes.Nature Biomedical Engineering, pages 1–20, 2025","work_id":"7d738772-f186-4fc4-8932-713b84087ee5","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":28,"snapshot_sha256":"aee1f08cf9eb0959da6783f1a542a33d5ea94c00e56beeefb4685a30f476bba1","internal_anchors":2},"formal_canon":{"evidence_count":2,"snapshot_sha256":"b765a28d2bbf49ff75779e59a90afcdca64ece329219acdf8f0b80e943937910"},"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":"2605.16104","created_at":"2026-05-20T00:01:52.844508+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.16104v1","created_at":"2026-05-20T00:01:52.844508+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.16104","created_at":"2026-05-20T00:01:52.844508+00:00"},{"alias_kind":"pith_short_12","alias_value":"PCECLMH6ZFNB","created_at":"2026-05-20T00:01:52.844508+00:00"},{"alias_kind":"pith_short_16","alias_value":"PCECLMH6ZFNBLTQ7","created_at":"2026-05-20T00:01:52.844508+00:00"},{"alias_kind":"pith_short_8","alias_value":"PCECLMH6","created_at":"2026-05-20T00:01:52.844508+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PCECLMH6ZFNBLTQ7HY77MOKVDB","json":"https://pith.science/pith/PCECLMH6ZFNBLTQ7HY77MOKVDB.json","graph_json":"https://pith.science/api/pith-number/PCECLMH6ZFNBLTQ7HY77MOKVDB/graph.json","events_json":"https://pith.science/api/pith-number/PCECLMH6ZFNBLTQ7HY77MOKVDB/events.json","paper":"https://pith.science/paper/PCECLMH6"},"agent_actions":{"view_html":"https://pith.science/pith/PCECLMH6ZFNBLTQ7HY77MOKVDB","download_json":"https://pith.science/pith/PCECLMH6ZFNBLTQ7HY77MOKVDB.json","view_paper":"https://pith.science/paper/PCECLMH6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.16104&json=true","fetch_graph":"https://pith.science/api/pith-number/PCECLMH6ZFNBLTQ7HY77MOKVDB/graph.json","fetch_events":"https://pith.science/api/pith-number/PCECLMH6ZFNBLTQ7HY77MOKVDB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PCECLMH6ZFNBLTQ7HY77MOKVDB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PCECLMH6ZFNBLTQ7HY77MOKVDB/action/storage_attestation","attest_author":"https://pith.science/pith/PCECLMH6ZFNBLTQ7HY77MOKVDB/action/author_attestation","sign_citation":"https://pith.science/pith/PCECLMH6ZFNBLTQ7HY77MOKVDB/action/citation_signature","submit_replication":"https://pith.science/pith/PCECLMH6ZFNBLTQ7HY77MOKVDB/action/replication_record"}},"created_at":"2026-05-20T00:01:52.844508+00:00","updated_at":"2026-05-20T00:01:52.844508+00:00"}