{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:S7MUHONQVFATVDTN2TUBG3EIWA","short_pith_number":"pith:S7MUHONQ","schema_version":"1.0","canonical_sha256":"97d943b9b0a9413a8e6dd4e8136c88b017a333d2b201c64a51bac99ff0de7fe3","source":{"kind":"arxiv","id":"1906.00295","version":1},"attestation_state":"computed","paper":{"title":"Multimodal Transformer for Unaligned Multimodal Language Sequences","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"J. Zico Kolter, Louis-Philippe Morency, Paul Pu Liang, Ruslan Salakhutdinov, Shaojie Bai, Yao-Hung Hubert Tsai","submitted_at":"2019-06-01T21:29:20Z","abstract_excerpt":"Human language is often multimodal, which comprehends a mixture of natural language, facial gestures, and acoustic behaviors. However, two major challenges in modeling such multimodal human language time-series data exist: 1) inherent data non-alignment due to variable sampling rates for the sequences from each modality; and 2) long-range dependencies between elements across modalities. In this paper, we introduce the Multimodal Transformer (MulT) to generically address the above issues in an end-to-end manner without explicitly aligning the data. At the heart of our model is the directional p"},"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":"1906.00295","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-06-01T21:29:20Z","cross_cats_sorted":[],"title_canon_sha256":"be90c46412fa54279758b26edbcd8443d497f4385b4b6f9e9fce3e32e12b0e46","abstract_canon_sha256":"20fc4b81822694c866c14432f94ac4ab8fa521bd60c0fa17ca0fa5475fdf0c85"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:44:27.536941Z","signature_b64":"HmREfLAG9EPo1UTfIJJLG6bcq+pbVmx9pHLppqNME01/krsxkOJRac8GG2yNpmlylb4JQsnidpF1WfvtONlMAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"97d943b9b0a9413a8e6dd4e8136c88b017a333d2b201c64a51bac99ff0de7fe3","last_reissued_at":"2026-05-17T23:44:27.536440Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:44:27.536440Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multimodal Transformer for Unaligned Multimodal Language Sequences","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"J. Zico Kolter, Louis-Philippe Morency, Paul Pu Liang, Ruslan Salakhutdinov, Shaojie Bai, Yao-Hung Hubert Tsai","submitted_at":"2019-06-01T21:29:20Z","abstract_excerpt":"Human language is often multimodal, which comprehends a mixture of natural language, facial gestures, and acoustic behaviors. However, two major challenges in modeling such multimodal human language time-series data exist: 1) inherent data non-alignment due to variable sampling rates for the sequences from each modality; and 2) long-range dependencies between elements across modalities. In this paper, we introduce the Multimodal Transformer (MulT) to generically address the above issues in an end-to-end manner without explicitly aligning the data. At the heart of our model is the directional p"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1906.00295","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":""},"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":"1906.00295","created_at":"2026-05-17T23:44:27.536522+00:00"},{"alias_kind":"arxiv_version","alias_value":"1906.00295v1","created_at":"2026-05-17T23:44:27.536522+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1906.00295","created_at":"2026-05-17T23:44:27.536522+00:00"},{"alias_kind":"pith_short_12","alias_value":"S7MUHONQVFAT","created_at":"2026-05-18T12:33:27.125529+00:00"},{"alias_kind":"pith_short_16","alias_value":"S7MUHONQVFATVDTN","created_at":"2026-05-18T12:33:27.125529+00:00"},{"alias_kind":"pith_short_8","alias_value":"S7MUHONQ","created_at":"2026-05-18T12:33:27.125529+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2509.09794","citing_title":"Synthetic Homes: A Multimodal Generative AI Pipeline for Residential Building Data Generation under Data Scarcity","ref_index":47,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/S7MUHONQVFATVDTN2TUBG3EIWA","json":"https://pith.science/pith/S7MUHONQVFATVDTN2TUBG3EIWA.json","graph_json":"https://pith.science/api/pith-number/S7MUHONQVFATVDTN2TUBG3EIWA/graph.json","events_json":"https://pith.science/api/pith-number/S7MUHONQVFATVDTN2TUBG3EIWA/events.json","paper":"https://pith.science/paper/S7MUHONQ"},"agent_actions":{"view_html":"https://pith.science/pith/S7MUHONQVFATVDTN2TUBG3EIWA","download_json":"https://pith.science/pith/S7MUHONQVFATVDTN2TUBG3EIWA.json","view_paper":"https://pith.science/paper/S7MUHONQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1906.00295&json=true","fetch_graph":"https://pith.science/api/pith-number/S7MUHONQVFATVDTN2TUBG3EIWA/graph.json","fetch_events":"https://pith.science/api/pith-number/S7MUHONQVFATVDTN2TUBG3EIWA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/S7MUHONQVFATVDTN2TUBG3EIWA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/S7MUHONQVFATVDTN2TUBG3EIWA/action/storage_attestation","attest_author":"https://pith.science/pith/S7MUHONQVFATVDTN2TUBG3EIWA/action/author_attestation","sign_citation":"https://pith.science/pith/S7MUHONQVFATVDTN2TUBG3EIWA/action/citation_signature","submit_replication":"https://pith.science/pith/S7MUHONQVFATVDTN2TUBG3EIWA/action/replication_record"}},"created_at":"2026-05-17T23:44:27.536522+00:00","updated_at":"2026-05-17T23:44:27.536522+00:00"}