{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:4ULQJ6ISN7Y6F7CMPXK4FUYYZR","short_pith_number":"pith:4ULQJ6IS","schema_version":"1.0","canonical_sha256":"e51704f9126ff1e2fc4c7dd5c2d318cc55309cbcf1e3a3b3d4f48d6e90c8fe20","source":{"kind":"arxiv","id":"2109.05125","version":1},"attestation_state":"computed","paper":{"title":"MURAL: Multimodal, Multitask Retrieval Across Languages","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.IR","authors_text":"Aashi Jain, Chao Jia, Jason Baldridge, Krishna Srinivasan, Mandy Guo, Sneha Kudugunta, Ting Chen, Yinfei Yang","submitted_at":"2021-09-10T22:26:05Z","abstract_excerpt":"Both image-caption pairs and translation pairs provide the means to learn deep representations of and connections between languages. We use both types of pairs in MURAL (MUltimodal, MUltitask Representations Across Languages), a dual encoder that solves two tasks: 1) image-text matching and 2) translation pair matching. By incorporating billions of translation pairs, MURAL extends ALIGN (Jia et al. PMLR'21)--a state-of-the-art dual encoder learned from 1.8 billion noisy image-text pairs. When using the same encoders, MURAL's performance matches or exceeds ALIGN's cross-modal retrieval performa"},"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":"2109.05125","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.IR","submitted_at":"2021-09-10T22:26:05Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"530b9f54461cdd0f2d4aa72459c5a4656252e49cb8d9e723474415f1e8c71ab6","abstract_canon_sha256":"bb79fa22dd12bbb5a8d59dd583f4f744574341ac81990e41ed2e0df482152961"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:13:34.915689Z","signature_b64":"yu6zGrUVXXShOutyRFjego6mIMxp6ngUZJqcTVCm5xzjwcmIIR+AQHxPkeYq39QE6h26Toni0iNbGkSRfDO9Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e51704f9126ff1e2fc4c7dd5c2d318cc55309cbcf1e3a3b3d4f48d6e90c8fe20","last_reissued_at":"2026-07-05T03:13:34.915228Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:13:34.915228Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MURAL: Multimodal, Multitask Retrieval Across Languages","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.IR","authors_text":"Aashi Jain, Chao Jia, Jason Baldridge, Krishna Srinivasan, Mandy Guo, Sneha Kudugunta, Ting Chen, Yinfei Yang","submitted_at":"2021-09-10T22:26:05Z","abstract_excerpt":"Both image-caption pairs and translation pairs provide the means to learn deep representations of and connections between languages. We use both types of pairs in MURAL (MUltimodal, MUltitask Representations Across Languages), a dual encoder that solves two tasks: 1) image-text matching and 2) translation pair matching. By incorporating billions of translation pairs, MURAL extends ALIGN (Jia et al. PMLR'21)--a state-of-the-art dual encoder learned from 1.8 billion noisy image-text pairs. When using the same encoders, MURAL's performance matches or exceeds ALIGN's cross-modal retrieval performa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2109.05125","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/2109.05125/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":"2109.05125","created_at":"2026-07-05T03:13:34.915279+00:00"},{"alias_kind":"arxiv_version","alias_value":"2109.05125v1","created_at":"2026-07-05T03:13:34.915279+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2109.05125","created_at":"2026-07-05T03:13:34.915279+00:00"},{"alias_kind":"pith_short_12","alias_value":"4ULQJ6ISN7Y6","created_at":"2026-07-05T03:13:34.915279+00:00"},{"alias_kind":"pith_short_16","alias_value":"4ULQJ6ISN7Y6F7CM","created_at":"2026-07-05T03:13:34.915279+00:00"},{"alias_kind":"pith_short_8","alias_value":"4ULQJ6IS","created_at":"2026-07-05T03:13:34.915279+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2509.22123","citing_title":"Multilingual Vision-Language Models, A Survey","ref_index":68,"is_internal_anchor":false},{"citing_arxiv_id":"2204.00598","citing_title":"Socratic Models: Composing Zero-Shot Multimodal Reasoning with Language","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2312.14238","citing_title":"InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Tasks","ref_index":70,"is_internal_anchor":false},{"citing_arxiv_id":"2204.14198","citing_title":"Flamingo: a Visual Language Model for Few-Shot Learning","ref_index":50,"is_internal_anchor":false},{"citing_arxiv_id":"2207.05608","citing_title":"Inner Monologue: Embodied Reasoning through Planning with Language Models","ref_index":71,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4ULQJ6ISN7Y6F7CMPXK4FUYYZR","json":"https://pith.science/pith/4ULQJ6ISN7Y6F7CMPXK4FUYYZR.json","graph_json":"https://pith.science/api/pith-number/4ULQJ6ISN7Y6F7CMPXK4FUYYZR/graph.json","events_json":"https://pith.science/api/pith-number/4ULQJ6ISN7Y6F7CMPXK4FUYYZR/events.json","paper":"https://pith.science/paper/4ULQJ6IS"},"agent_actions":{"view_html":"https://pith.science/pith/4ULQJ6ISN7Y6F7CMPXK4FUYYZR","download_json":"https://pith.science/pith/4ULQJ6ISN7Y6F7CMPXK4FUYYZR.json","view_paper":"https://pith.science/paper/4ULQJ6IS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2109.05125&json=true","fetch_graph":"https://pith.science/api/pith-number/4ULQJ6ISN7Y6F7CMPXK4FUYYZR/graph.json","fetch_events":"https://pith.science/api/pith-number/4ULQJ6ISN7Y6F7CMPXK4FUYYZR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4ULQJ6ISN7Y6F7CMPXK4FUYYZR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4ULQJ6ISN7Y6F7CMPXK4FUYYZR/action/storage_attestation","attest_author":"https://pith.science/pith/4ULQJ6ISN7Y6F7CMPXK4FUYYZR/action/author_attestation","sign_citation":"https://pith.science/pith/4ULQJ6ISN7Y6F7CMPXK4FUYYZR/action/citation_signature","submit_replication":"https://pith.science/pith/4ULQJ6ISN7Y6F7CMPXK4FUYYZR/action/replication_record"}},"created_at":"2026-07-05T03:13:34.915279+00:00","updated_at":"2026-07-05T03:13:34.915279+00:00"}