{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:BO5ZETSEIKC3HJBXSZMU7AHVFP","short_pith_number":"pith:BO5ZETSE","schema_version":"1.0","canonical_sha256":"0bbb924e444285b3a43796594f80f52bca20f42ce78e9b7c89c59e734e2c218e","source":{"kind":"arxiv","id":"2310.13267","version":1},"attestation_state":"computed","paper":{"title":"On the Language Encoder of Contrastive Cross-modal Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV","cs.LG","cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Chieh-Hsin Lai, Hiromi Wakaki, Junya Ono, Mengjie Zhao, Naoki Murata, Takashi Shibuya, Wei-Hsiang Liao, Yuhta Takida, Yuki Mitsufuji, Zhi Zhong","submitted_at":"2023-10-20T04:21:09Z","abstract_excerpt":"Contrastive cross-modal models such as CLIP and CLAP aid various vision-language (VL) and audio-language (AL) tasks. However, there has been limited investigation of and improvement in their language encoder, which is the central component of encoding natural language descriptions of image/audio into vector representations. We extensively evaluate how unsupervised and supervised sentence embedding training affect language encoder quality and cross-modal task performance. In VL pretraining, we found that sentence embedding training language encoder quality and aids in cross-modal tasks, improvi"},"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":"2310.13267","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-20T04:21:09Z","cross_cats_sorted":["cs.CV","cs.LG","cs.SD","eess.AS"],"title_canon_sha256":"4b498833ca55b76671d131d392c66327a7be6d45a24608420a06ba9687351a05","abstract_canon_sha256":"2d0f8694d384e55b61346c677e7d1ee152b9a450356ca79fcfed7fcd41037376"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:03:00.332356Z","signature_b64":"sXjoZTyM84a9/8MTmypnnzQ8FWG0xSu8K8Zn3p/UicIH8SbcB/kiYpM/MXxYRx/PmcPohASb7PrawuK9d3xdBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0bbb924e444285b3a43796594f80f52bca20f42ce78e9b7c89c59e734e2c218e","last_reissued_at":"2026-07-05T07:03:00.331882Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:03:00.331882Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On the Language Encoder of Contrastive Cross-modal Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV","cs.LG","cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Chieh-Hsin Lai, Hiromi Wakaki, Junya Ono, Mengjie Zhao, Naoki Murata, Takashi Shibuya, Wei-Hsiang Liao, Yuhta Takida, Yuki Mitsufuji, Zhi Zhong","submitted_at":"2023-10-20T04:21:09Z","abstract_excerpt":"Contrastive cross-modal models such as CLIP and CLAP aid various vision-language (VL) and audio-language (AL) tasks. However, there has been limited investigation of and improvement in their language encoder, which is the central component of encoding natural language descriptions of image/audio into vector representations. We extensively evaluate how unsupervised and supervised sentence embedding training affect language encoder quality and cross-modal task performance. In VL pretraining, we found that sentence embedding training language encoder quality and aids in cross-modal tasks, improvi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.13267","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/2310.13267/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":"2310.13267","created_at":"2026-07-05T07:03:00.331938+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.13267v1","created_at":"2026-07-05T07:03:00.331938+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.13267","created_at":"2026-07-05T07:03:00.331938+00:00"},{"alias_kind":"pith_short_12","alias_value":"BO5ZETSEIKC3","created_at":"2026-07-05T07:03:00.331938+00:00"},{"alias_kind":"pith_short_16","alias_value":"BO5ZETSEIKC3HJBX","created_at":"2026-07-05T07:03:00.331938+00:00"},{"alias_kind":"pith_short_8","alias_value":"BO5ZETSE","created_at":"2026-07-05T07:03:00.331938+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/BO5ZETSEIKC3HJBXSZMU7AHVFP","json":"https://pith.science/pith/BO5ZETSEIKC3HJBXSZMU7AHVFP.json","graph_json":"https://pith.science/api/pith-number/BO5ZETSEIKC3HJBXSZMU7AHVFP/graph.json","events_json":"https://pith.science/api/pith-number/BO5ZETSEIKC3HJBXSZMU7AHVFP/events.json","paper":"https://pith.science/paper/BO5ZETSE"},"agent_actions":{"view_html":"https://pith.science/pith/BO5ZETSEIKC3HJBXSZMU7AHVFP","download_json":"https://pith.science/pith/BO5ZETSEIKC3HJBXSZMU7AHVFP.json","view_paper":"https://pith.science/paper/BO5ZETSE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.13267&json=true","fetch_graph":"https://pith.science/api/pith-number/BO5ZETSEIKC3HJBXSZMU7AHVFP/graph.json","fetch_events":"https://pith.science/api/pith-number/BO5ZETSEIKC3HJBXSZMU7AHVFP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BO5ZETSEIKC3HJBXSZMU7AHVFP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BO5ZETSEIKC3HJBXSZMU7AHVFP/action/storage_attestation","attest_author":"https://pith.science/pith/BO5ZETSEIKC3HJBXSZMU7AHVFP/action/author_attestation","sign_citation":"https://pith.science/pith/BO5ZETSEIKC3HJBXSZMU7AHVFP/action/citation_signature","submit_replication":"https://pith.science/pith/BO5ZETSEIKC3HJBXSZMU7AHVFP/action/replication_record"}},"created_at":"2026-07-05T07:03:00.331938+00:00","updated_at":"2026-07-05T07:03:00.331938+00:00"}