{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:XUL5CWWBB6WZWSJIYOOA7ZDPYI","short_pith_number":"pith:XUL5CWWB","schema_version":"1.0","canonical_sha256":"bd17d15ac10fad9b4928c39c0fe46fc23a2d77466274b1177b7132453c90dd08","source":{"kind":"arxiv","id":"2103.09666","version":3},"attestation_state":"computed","paper":{"title":"Multimodal End-to-End Sparse Model for Emotion Recognition","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Pascale Fung, Samuel Cahyawijaya, Wenliang Dai, Zihan Liu","submitted_at":"2021-03-17T14:05:05Z","abstract_excerpt":"Existing works on multimodal affective computing tasks, such as emotion recognition, generally adopt a two-phase pipeline, first extracting feature representations for each single modality with hand-crafted algorithms and then performing end-to-end learning with the extracted features. However, the extracted features are fixed and cannot be further fine-tuned on different target tasks, and manually finding feature extraction algorithms does not generalize or scale well to different tasks, which can lead to sub-optimal performance. In this paper, we develop a fully end-to-end model that connect"},"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":"2103.09666","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-03-17T14:05:05Z","cross_cats_sorted":[],"title_canon_sha256":"46f31d2871e919ac98da18de324ffc3d1e73f6753aa0f069df2a554f1bc8b7fe","abstract_canon_sha256":"bdba125259e2399579064e6e4336a996099d39f2f7cf80ed452990eb7e2a84d4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:37:25.514480Z","signature_b64":"v8nTEeI55bGARHy+GVL4PaL3y82xhGvVVqzbyiY+uQT1B9GMCWHqsBNg3XeiakVH5Ltjr4qR6oJ/dxOa1tOCDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bd17d15ac10fad9b4928c39c0fe46fc23a2d77466274b1177b7132453c90dd08","last_reissued_at":"2026-07-05T03:37:25.514012Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:37:25.514012Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multimodal End-to-End Sparse Model for Emotion Recognition","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Pascale Fung, Samuel Cahyawijaya, Wenliang Dai, Zihan Liu","submitted_at":"2021-03-17T14:05:05Z","abstract_excerpt":"Existing works on multimodal affective computing tasks, such as emotion recognition, generally adopt a two-phase pipeline, first extracting feature representations for each single modality with hand-crafted algorithms and then performing end-to-end learning with the extracted features. However, the extracted features are fixed and cannot be further fine-tuned on different target tasks, and manually finding feature extraction algorithms does not generalize or scale well to different tasks, which can lead to sub-optimal performance. In this paper, we develop a fully end-to-end model that connect"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2103.09666","kind":"arxiv","version":3},"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/2103.09666/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":"2103.09666","created_at":"2026-07-05T03:37:25.514075+00:00"},{"alias_kind":"arxiv_version","alias_value":"2103.09666v3","created_at":"2026-07-05T03:37:25.514075+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2103.09666","created_at":"2026-07-05T03:37:25.514075+00:00"},{"alias_kind":"pith_short_12","alias_value":"XUL5CWWBB6WZ","created_at":"2026-07-05T03:37:25.514075+00:00"},{"alias_kind":"pith_short_16","alias_value":"XUL5CWWBB6WZWSJI","created_at":"2026-07-05T03:37:25.514075+00:00"},{"alias_kind":"pith_short_8","alias_value":"XUL5CWWB","created_at":"2026-07-05T03:37:25.514075+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.18568","citing_title":"Learning without Isolation: Pathway Protection for Continual Learning","ref_index":2013,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XUL5CWWBB6WZWSJIYOOA7ZDPYI","json":"https://pith.science/pith/XUL5CWWBB6WZWSJIYOOA7ZDPYI.json","graph_json":"https://pith.science/api/pith-number/XUL5CWWBB6WZWSJIYOOA7ZDPYI/graph.json","events_json":"https://pith.science/api/pith-number/XUL5CWWBB6WZWSJIYOOA7ZDPYI/events.json","paper":"https://pith.science/paper/XUL5CWWB"},"agent_actions":{"view_html":"https://pith.science/pith/XUL5CWWBB6WZWSJIYOOA7ZDPYI","download_json":"https://pith.science/pith/XUL5CWWBB6WZWSJIYOOA7ZDPYI.json","view_paper":"https://pith.science/paper/XUL5CWWB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2103.09666&json=true","fetch_graph":"https://pith.science/api/pith-number/XUL5CWWBB6WZWSJIYOOA7ZDPYI/graph.json","fetch_events":"https://pith.science/api/pith-number/XUL5CWWBB6WZWSJIYOOA7ZDPYI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XUL5CWWBB6WZWSJIYOOA7ZDPYI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XUL5CWWBB6WZWSJIYOOA7ZDPYI/action/storage_attestation","attest_author":"https://pith.science/pith/XUL5CWWBB6WZWSJIYOOA7ZDPYI/action/author_attestation","sign_citation":"https://pith.science/pith/XUL5CWWBB6WZWSJIYOOA7ZDPYI/action/citation_signature","submit_replication":"https://pith.science/pith/XUL5CWWBB6WZWSJIYOOA7ZDPYI/action/replication_record"}},"created_at":"2026-07-05T03:37:25.514075+00:00","updated_at":"2026-07-05T03:37:25.514075+00:00"}