{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:ABAV7L6P52VEXA42VNQJAGDAVO","short_pith_number":"pith:ABAV7L6P","schema_version":"1.0","canonical_sha256":"00415fafcfeeaa4b839aab60901860aba60ba006bbaf99f5cb64e61aa65536e1","source":{"kind":"arxiv","id":"2606.05713","version":1},"attestation_state":"computed","paper":{"title":"Beyond Generative Decoding: Discriminative Hidden-State Readout from a Native Omni-Modal LLM for Multimodal Sentiment Analysis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.MM","authors_text":"Bin Wen, Tien-Ping Tan","submitted_at":"2026-06-04T05:12:36Z","abstract_excerpt":"Multimodal sentiment analysis (MSA) infers human affect from language, acoustic, and visual signals. Recent methods increasingly adapt large multimodal models (LMMs) via generative readout: prompting the model to emit a sentiment score as a text string. While convenient, this ties continuous regression to discrete autoregressive decoding, incurring unmeasured costs. We revisit this readout mechanism and propose a discriminative formulation built on the Thinker module of a native omni-modal LLM (Qwen2.5-Omni-7B). Instead of text decoding, we map the final-layer hidden state of the last non-padd"},"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":"2606.05713","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.MM","submitted_at":"2026-06-04T05:12:36Z","cross_cats_sorted":["cs.SD","eess.AS"],"title_canon_sha256":"1b5079d704b17c586ccff817d09febe40ebd16aa3172a3e7ba906b75e3df9c81","abstract_canon_sha256":"b803fc470b97e56a0d04706497fedc770ce0e421e7f1d68177c6ec6b6ccd5f1c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-05T01:15:00.054592Z","signature_b64":"JRxFz5+B8frEZYp+aktaZHNWtdQchPPkJXyXPLd7rehTWXCGJQvtpOsTKn+rukmei95EqFVo5yxJtMxiFHZyBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"00415fafcfeeaa4b839aab60901860aba60ba006bbaf99f5cb64e61aa65536e1","last_reissued_at":"2026-06-05T01:15:00.054146Z","signature_status":"signed_v1","first_computed_at":"2026-06-05T01:15:00.054146Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Beyond Generative Decoding: Discriminative Hidden-State Readout from a Native Omni-Modal LLM for Multimodal Sentiment Analysis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.MM","authors_text":"Bin Wen, Tien-Ping Tan","submitted_at":"2026-06-04T05:12:36Z","abstract_excerpt":"Multimodal sentiment analysis (MSA) infers human affect from language, acoustic, and visual signals. Recent methods increasingly adapt large multimodal models (LMMs) via generative readout: prompting the model to emit a sentiment score as a text string. While convenient, this ties continuous regression to discrete autoregressive decoding, incurring unmeasured costs. We revisit this readout mechanism and propose a discriminative formulation built on the Thinker module of a native omni-modal LLM (Qwen2.5-Omni-7B). Instead of text decoding, we map the final-layer hidden state of the last non-padd"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.05713","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/2606.05713/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":"2606.05713","created_at":"2026-06-05T01:15:00.054205+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.05713v1","created_at":"2026-06-05T01:15:00.054205+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.05713","created_at":"2026-06-05T01:15:00.054205+00:00"},{"alias_kind":"pith_short_12","alias_value":"ABAV7L6P52VE","created_at":"2026-06-05T01:15:00.054205+00:00"},{"alias_kind":"pith_short_16","alias_value":"ABAV7L6P52VEXA42","created_at":"2026-06-05T01:15:00.054205+00:00"},{"alias_kind":"pith_short_8","alias_value":"ABAV7L6P","created_at":"2026-06-05T01:15:00.054205+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/ABAV7L6P52VEXA42VNQJAGDAVO","json":"https://pith.science/pith/ABAV7L6P52VEXA42VNQJAGDAVO.json","graph_json":"https://pith.science/api/pith-number/ABAV7L6P52VEXA42VNQJAGDAVO/graph.json","events_json":"https://pith.science/api/pith-number/ABAV7L6P52VEXA42VNQJAGDAVO/events.json","paper":"https://pith.science/paper/ABAV7L6P"},"agent_actions":{"view_html":"https://pith.science/pith/ABAV7L6P52VEXA42VNQJAGDAVO","download_json":"https://pith.science/pith/ABAV7L6P52VEXA42VNQJAGDAVO.json","view_paper":"https://pith.science/paper/ABAV7L6P","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.05713&json=true","fetch_graph":"https://pith.science/api/pith-number/ABAV7L6P52VEXA42VNQJAGDAVO/graph.json","fetch_events":"https://pith.science/api/pith-number/ABAV7L6P52VEXA42VNQJAGDAVO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ABAV7L6P52VEXA42VNQJAGDAVO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ABAV7L6P52VEXA42VNQJAGDAVO/action/storage_attestation","attest_author":"https://pith.science/pith/ABAV7L6P52VEXA42VNQJAGDAVO/action/author_attestation","sign_citation":"https://pith.science/pith/ABAV7L6P52VEXA42VNQJAGDAVO/action/citation_signature","submit_replication":"https://pith.science/pith/ABAV7L6P52VEXA42VNQJAGDAVO/action/replication_record"}},"created_at":"2026-06-05T01:15:00.054205+00:00","updated_at":"2026-06-05T01:15:00.054205+00:00"}