{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:IDAIXY2P3HBLYFHQJNLZPRFUK5","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":"f8b58a6f628877e37c734900381d7d3d0fcfc4688b6e8314c5c2e43847add592","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-16T09:03:31Z","title_canon_sha256":"e852d903a3cd969ef7979f5761b8b1bb2b12d14f0df5653012166d222a2499d2"},"schema_version":"1.0","source":{"id":"2605.16889","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.16889","created_at":"2026-05-20T00:03:28Z"},{"alias_kind":"arxiv_version","alias_value":"2605.16889v1","created_at":"2026-05-20T00:03:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.16889","created_at":"2026-05-20T00:03:28Z"},{"alias_kind":"pith_short_12","alias_value":"IDAIXY2P3HBL","created_at":"2026-05-20T00:03:28Z"},{"alias_kind":"pith_short_16","alias_value":"IDAIXY2P3HBLYFHQ","created_at":"2026-05-20T00:03:28Z"},{"alias_kind":"pith_short_8","alias_value":"IDAIXY2P","created_at":"2026-05-20T00:03:28Z"}],"graph_snapshots":[{"event_id":"sha256:afff5fb71440a53495a436d375097ad0828d4647bf011cde4914aabd2d4cb070","target":"graph","created_at":"2026-05-20T00:03:28Z","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 framework maintains stable and reliable sentiment predictions under diverse missing-modality patterns, achieving state-of-the-art performance on CMU-MOSI and CMU-MOSEI with ACC of 86.28% and 85.88%, and F1 of 86.24% and 85.86% under full-modality input."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That complete-modality samples provide sufficiently stable references to constrain representations and align different modality combinations into a shared sentiment space without introducing new distribution shifts."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"A two-level reference alignment framework uses complete-modality samples and prototype voting to reduce decision drift and improve robustness in multimodal sentiment analysis under missing modalities."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"A two-level reference alignment framework maintains stable sentiment predictions under missing modalities by anchoring both features and decisions to complete samples."}],"snapshot_sha256":"e620d543160bae9e6ec11dcb20ad9c8126369419a666171ac951ddaae5e1776e"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"e8e9756b3c9f9f1c2a8ab9b655b80d6e40225a9de609903f77eb2f24827d3e4e"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T22:01:19.557882Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T21:50:54.961665Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"cited_work_retraction","ran_at":"2026-05-19T20:52:41.924540Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-19T18:41:56.284714Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T18:33:26.362769Z","status":"skipped","version":"1.0.0"}],"endpoint":"/pith/2605.16889/integrity.json","findings":[],"snapshot_sha256":"2be5d693cba17ea2b7ab5dd97611b6dae31407a6d24844d8386c814dabf0110a","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Multimodal sentiment analysis relies on textual, acoustic, and visual signals, yet real-world data often suffer from modality missing and quality imbalance. Existing methods generate features for modality missing from available ones, but differences in expression mechanisms and sentiment dynamics across modalities may cause the generated features to deviate from true distributions and mislead prediction. In addition, unreliable modalities may dominate fusion, resulting in representation shift across modality combinations and unstable sentiment representations. To address these challenges, we p","authors_text":"Chenglizhao Chen, Guisheng Zhang, Mengke Song, Xiaomin Yu, Xinyu Liu, Yuchen Cao","cross_cats":[],"headline":"A two-level reference alignment framework maintains stable sentiment predictions under missing modalities by anchoring both features and decisions to complete samples.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-16T09:03:31Z","title":"Controlling Decision Drift in Multimodal Sentiment Analysis with Missing Modalities"},"references":{"count":36,"internal_anchors":5,"resolved_work":36,"sample":[{"cited_arxiv_id":"1702.01992","doi":"","is_internal_anchor":true,"ref_index":1,"title":"Gated Multimodal Units for Information Fusion","work_id":"da7f3da5-ca1f-4da5-bbee-fc195ced03ba","year":2017},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Openface: an open source facial behavior analysis toolkit","work_id":"864d67e9-e832-44bd-bbd6-9b7b2c541c9c","year":2016},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Multimodal ma- chine learning: A survey and taxonomy.IEEE TPAMI, 41(2):423–443,","work_id":"9bf52045-fefd-46c3-9ec3-e80c94829452","year":2018},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Ucmib-pns: Balancing sufficiency and ne- cessity with probabilistic causality and cross-modal uncer- tainty in multimodal sentiment analysis.IEEE TAC,","work_id":"f569ee18-3612-4216-9367-879e70637ebb","year":2025},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Unbiased missing-modality mul- timodal learning","work_id":"a801365e-af5c-4c3b-9720-72d44e1e116d","year":2025}],"snapshot_sha256":"c5692cbcc4e697c1c4d76abb4d781a20b0cf136e158adfd62bba3e20dd686b88"},"source":{"id":"2605.16889","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-19T21:41:57.460178Z","id":"a124fb7e-690d-4e9b-841c-cb8867df7c05","model_set":{"reader":"grok-4.3"},"one_line_summary":"A two-level reference alignment framework uses complete-modality samples and prototype voting to reduce decision drift and improve robustness in multimodal sentiment analysis under missing modalities.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"A two-level reference alignment framework maintains stable sentiment predictions under missing modalities by anchoring both features and decisions to complete samples.","strongest_claim":"The framework maintains stable and reliable sentiment predictions under diverse missing-modality patterns, achieving state-of-the-art performance on CMU-MOSI and CMU-MOSEI with ACC of 86.28% and 85.88%, and F1 of 86.24% and 85.86% under full-modality input.","weakest_assumption":"That complete-modality samples provide sufficiently stable references to constrain representations and align different modality combinations into a shared sentiment space without introducing new distribution shifts."}},"verdict_id":"a124fb7e-690d-4e9b-841c-cb8867df7c05"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:cde58e18b17dfefa56eeae1924bc8cdabd3b843caf3ebadc13ce4e596665c2bc","target":"record","created_at":"2026-05-20T00:03:28Z","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":"f8b58a6f628877e37c734900381d7d3d0fcfc4688b6e8314c5c2e43847add592","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-16T09:03:31Z","title_canon_sha256":"e852d903a3cd969ef7979f5761b8b1bb2b12d14f0df5653012166d222a2499d2"},"schema_version":"1.0","source":{"id":"2605.16889","kind":"arxiv","version":1}},"canonical_sha256":"40c08be34fd9c2bc14f04b5797c4b457790b055695de415e10196d567cee5012","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"40c08be34fd9c2bc14f04b5797c4b457790b055695de415e10196d567cee5012","first_computed_at":"2026-05-20T00:03:28.537818Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:03:28.537818Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"63Un/RKqli2u+6g/6p3GvhkfF3ghcFuALQSXyuRmUVbfjfxZtqIZrp+eS6LTYNwSHv0cGNzXW5VlqugZMgAzBg==","signature_status":"signed_v1","signed_at":"2026-05-20T00:03:28.538543Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.16889","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:cde58e18b17dfefa56eeae1924bc8cdabd3b843caf3ebadc13ce4e596665c2bc","sha256:afff5fb71440a53495a436d375097ad0828d4647bf011cde4914aabd2d4cb070"],"state_sha256":"e129e85beca2540ae0f593090570fc5a7c67dab3411b53ad410028ab58e57079"}