{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:4PRMUGOXSYRWGR3POM7UAUBIO2","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":"63a8801f3385b7f89066309346fc4efe2d69bd7fa8c5d3ff52ecdeeae840a831","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-13T19:32:11Z","title_canon_sha256":"73214917211d846889c2eedb50f229c1143ce1e33b627ce0ef3f22fda11508fc"},"schema_version":"1.0","source":{"id":"2605.15235","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.15235","created_at":"2026-05-20T00:00:47Z"},{"alias_kind":"arxiv_version","alias_value":"2605.15235v1","created_at":"2026-05-20T00:00:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.15235","created_at":"2026-05-20T00:00:47Z"},{"alias_kind":"pith_short_12","alias_value":"4PRMUGOXSYRW","created_at":"2026-05-20T00:00:47Z"},{"alias_kind":"pith_short_16","alias_value":"4PRMUGOXSYRWGR3P","created_at":"2026-05-20T00:00:47Z"},{"alias_kind":"pith_short_8","alias_value":"4PRMUGOX","created_at":"2026-05-20T00:00:47Z"}],"graph_snapshots":[{"event_id":"sha256:6cbb4f6bf3fa9ee4af1becd76983021431bc6745357fed179c9794d77d89168f","target":"graph","created_at":"2026-05-20T00:00:47Z","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":"architecture family is the strongest predictor of robustness, outweighing parameter count"},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The nine chosen clinical datasets and six fusion architectures sufficiently represent the range of real-world multimodal physiological signals and sensor-failure patterns so that the observed robustness rankings generalize beyond the tested cases. (Abstract)"},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"MuteBench evaluates multimodal fusion robustness to modality missing and within-modality missing on 125000 samples from 9 clinical datasets, finding architecture family predicts tolerance better than parameter count."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Architecture family predicts robustness to missing modalities better than model size in clinical fusion."}],"snapshot_sha256":"423ce505b8717e52ff4103d1cebd7ab21cf6c4474081c58f53e9f98b17a5c526"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-19T17:41:56.139484Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T17:01:18.468993Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T16:36:43.039369Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T13:33:22.826836Z","status":"skipped","version":"1.0.0"}],"endpoint":"/pith/2605.15235/integrity.json","findings":[],"snapshot_sha256":"86864cf054fc1f6e7ae938b499a98e38af15e90dd58d53609545cf7b305efaa4","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Multimodal physiological data powers clinical AI systems from intensive care units to wearable devices, but sensors routinely fail in practice. Two failure modes are common: modality missing, where an entire channel is absent, and within-modality missing, where a contiguous time segment is lost. No existing benchmark evaluates multiple fusion architectures under both failure modes at controlled severity levels across diverse clinical datasets. We present MuteBench, a benchmark covering 9 datasets from 7 clinical domains, 6 fusion architectures, and 2 missing-data modes over 125,000 samples. Th","authors_text":"Chen Chen, Song Wang, Tianlong Chen, Wugeng Zheng, Ziwen Kan","cross_cats":[],"headline":"Architecture family predicts robustness to missing modalities better than model size in clinical fusion.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-13T19:32:11Z","title":"MuteBench: Modality Unavailability Tolerance Evaluation for Incomplete Multimodal Fusion"},"references":{"count":63,"internal_anchors":1,"resolved_work":63,"sample":[{"cited_arxiv_id":"","doi":"10.1038/s41598-026-39035-z","is_internal_anchor":false,"ref_index":1,"title":"doi: 10.1038/s41598-026-39035-z","work_id":"9837de1e-f997-48bf-941b-e515f83283c0","year":2026},{"cited_arxiv_id":"","doi":"10.1038/s41591-022-01981-2","is_internal_anchor":false,"ref_index":2,"title":"Multimodal biomedical AI.Nature Medicine, 28(9):1773–1784, 2022","work_id":"8658170a-8b8a-4db9-877f-2021053de5e4","year":2022},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Ehrxqa: A multi-modal ques- tion answering dataset for electronic health records with chest x-ray images","work_id":"859bc897-b6db-4296-a896-c29001d8e3d2","year":2023},{"cited_arxiv_id":"","doi":"10.1145/1553374.1553380","is_internal_anchor":false,"ref_index":4,"title":"In Proceedings of the 26th Annual International Conference on Machine Learning (Montreal, Quebec, Canada) (ICML ’09)","work_id":"5d30e737-af53-4f1d-8f90-c8e0159a22ba","year":2009},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Recurrent neural networks for multivariate time series with missing values.Scientific Reports, 8(1):6085,","work_id":"c7da1c42-db7d-409c-b51a-f2a527136b5e","year":null}],"snapshot_sha256":"1e19b346919c5010097f620cf0de7911d5d0e3c42c4b08281722e1a9176dc765"},"source":{"id":"2605.15235","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-19T16:29:45.622394Z","id":"f17af8ad-701b-4dd4-a586-b674601037af","model_set":{"reader":"grok-4.3"},"one_line_summary":"MuteBench evaluates multimodal fusion robustness to modality missing and within-modality missing on 125000 samples from 9 clinical datasets, finding architecture family predicts tolerance better than parameter count.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Architecture family predicts robustness to missing modalities better than model size in clinical fusion.","strongest_claim":"architecture family is the strongest predictor of robustness, outweighing parameter count","weakest_assumption":"The nine chosen clinical datasets and six fusion architectures sufficiently represent the range of real-world multimodal physiological signals and sensor-failure patterns so that the observed robustness rankings generalize beyond the tested cases. (Abstract)"}},"verdict_id":"f17af8ad-701b-4dd4-a586-b674601037af"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:1913956e715ed7af404d553a4a113dc7b6a88264dec3a3f39bba7308afdc946a","target":"record","created_at":"2026-05-20T00:00:47Z","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":"63a8801f3385b7f89066309346fc4efe2d69bd7fa8c5d3ff52ecdeeae840a831","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-13T19:32:11Z","title_canon_sha256":"73214917211d846889c2eedb50f229c1143ce1e33b627ce0ef3f22fda11508fc"},"schema_version":"1.0","source":{"id":"2605.15235","kind":"arxiv","version":1}},"canonical_sha256":"e3e2ca19d7962363476f733f40502876b57593779c266683967926c9054589b8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e3e2ca19d7962363476f733f40502876b57593779c266683967926c9054589b8","first_computed_at":"2026-05-20T00:00:47.719823Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:00:47.719823Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"NtkpG8226w45WuiVnJ5Xd9miV43+byN8hWwmM15L66TQ0C6EvqaY8VPj4+PAOZtR9+/7UYdBb6rctISwotb6CA==","signature_status":"signed_v1","signed_at":"2026-05-20T00:00:47.720758Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.15235","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1913956e715ed7af404d553a4a113dc7b6a88264dec3a3f39bba7308afdc946a","sha256:6cbb4f6bf3fa9ee4af1becd76983021431bc6745357fed179c9794d77d89168f"],"state_sha256":"f0e5ae03e0af88fa528e69cd7c81240ef2dc1681a56fbc00570b8651dce8921e"}