{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:YKDPZH3ES5WHR2XCSCSFTDFFSJ","short_pith_number":"pith:YKDPZH3E","schema_version":"1.0","canonical_sha256":"c286fc9f64976c78eae290a4598ca592558fac85b18ccc3641bcbe8f3d4591a3","source":{"kind":"arxiv","id":"1704.00260","version":2},"attestation_state":"computed","paper":{"title":"Aligned Image-Word Representations Improve Inductive Transfer Across Vision-Language Tasks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.NE","stat.ML"],"primary_cat":"cs.CV","authors_text":"Derek Hoiem, Kevin Shih, Saurabh Singh, Tanmay Gupta","submitted_at":"2017-04-02T08:01:30Z","abstract_excerpt":"An important goal of computer vision is to build systems that learn visual representations over time that can be applied to many tasks. In this paper, we investigate a vision-language embedding as a core representation and show that it leads to better cross-task transfer than standard multi-task learning. In particular, the task of visual recognition is aligned to the task of visual question answering by forcing each to use the same word-region embeddings. We show this leads to greater inductive transfer from recognition to VQA than standard multitask learning. Visual recognition also improves"},"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":"1704.00260","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2017-04-02T08:01:30Z","cross_cats_sorted":["cs.AI","cs.LG","cs.NE","stat.ML"],"title_canon_sha256":"c39260a660f9f25d9d52d9f0deb54b1eab5fe17eaf797fcedd12b27362cbe119","abstract_canon_sha256":"3c2d40458fb65665c0e1a997ce878cf3de11f0e634eb5c4575c47ad80b7ee723"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:32:50.924902Z","signature_b64":"6Oeu+hckP2nhY87LE7gWl5BO8r+rZ1VfLfh/qYqWER5NMnLvVkL17Oz/1FfrvhEbrxYaaRffq2/4t58pl2qaBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c286fc9f64976c78eae290a4598ca592558fac85b18ccc3641bcbe8f3d4591a3","last_reissued_at":"2026-05-18T00:32:50.924256Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:32:50.924256Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Aligned Image-Word Representations Improve Inductive Transfer Across Vision-Language Tasks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.NE","stat.ML"],"primary_cat":"cs.CV","authors_text":"Derek Hoiem, Kevin Shih, Saurabh Singh, Tanmay Gupta","submitted_at":"2017-04-02T08:01:30Z","abstract_excerpt":"An important goal of computer vision is to build systems that learn visual representations over time that can be applied to many tasks. In this paper, we investigate a vision-language embedding as a core representation and show that it leads to better cross-task transfer than standard multi-task learning. In particular, the task of visual recognition is aligned to the task of visual question answering by forcing each to use the same word-region embeddings. We show this leads to greater inductive transfer from recognition to VQA than standard multitask learning. Visual recognition also improves"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1704.00260","kind":"arxiv","version":2},"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":"1704.00260","created_at":"2026-05-18T00:32:50.924360+00:00"},{"alias_kind":"arxiv_version","alias_value":"1704.00260v2","created_at":"2026-05-18T00:32:50.924360+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1704.00260","created_at":"2026-05-18T00:32:50.924360+00:00"},{"alias_kind":"pith_short_12","alias_value":"YKDPZH3ES5WH","created_at":"2026-05-18T12:31:56.362134+00:00"},{"alias_kind":"pith_short_16","alias_value":"YKDPZH3ES5WHR2XC","created_at":"2026-05-18T12:31:56.362134+00:00"},{"alias_kind":"pith_short_8","alias_value":"YKDPZH3E","created_at":"2026-05-18T12:31:56.362134+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/YKDPZH3ES5WHR2XCSCSFTDFFSJ","json":"https://pith.science/pith/YKDPZH3ES5WHR2XCSCSFTDFFSJ.json","graph_json":"https://pith.science/api/pith-number/YKDPZH3ES5WHR2XCSCSFTDFFSJ/graph.json","events_json":"https://pith.science/api/pith-number/YKDPZH3ES5WHR2XCSCSFTDFFSJ/events.json","paper":"https://pith.science/paper/YKDPZH3E"},"agent_actions":{"view_html":"https://pith.science/pith/YKDPZH3ES5WHR2XCSCSFTDFFSJ","download_json":"https://pith.science/pith/YKDPZH3ES5WHR2XCSCSFTDFFSJ.json","view_paper":"https://pith.science/paper/YKDPZH3E","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1704.00260&json=true","fetch_graph":"https://pith.science/api/pith-number/YKDPZH3ES5WHR2XCSCSFTDFFSJ/graph.json","fetch_events":"https://pith.science/api/pith-number/YKDPZH3ES5WHR2XCSCSFTDFFSJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YKDPZH3ES5WHR2XCSCSFTDFFSJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YKDPZH3ES5WHR2XCSCSFTDFFSJ/action/storage_attestation","attest_author":"https://pith.science/pith/YKDPZH3ES5WHR2XCSCSFTDFFSJ/action/author_attestation","sign_citation":"https://pith.science/pith/YKDPZH3ES5WHR2XCSCSFTDFFSJ/action/citation_signature","submit_replication":"https://pith.science/pith/YKDPZH3ES5WHR2XCSCSFTDFFSJ/action/replication_record"}},"created_at":"2026-05-18T00:32:50.924360+00:00","updated_at":"2026-05-18T00:32:50.924360+00:00"}