{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:UTJQLXLH3RFKVA3AGQWEKD4LGG","short_pith_number":"pith:UTJQLXLH","schema_version":"1.0","canonical_sha256":"a4d305dd67dc4aaa8360342c450f8b31817e108f6c780f6b53e888541e398391","source":{"kind":"arxiv","id":"2010.03644","version":1},"attestation_state":"computed","paper":{"title":"Towards Understanding Sample Variance in Visually Grounded Language Generation: Evaluations and Observations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.CL","authors_text":"Kazoo Sone, Pradyumna Narayana, Sugato Basu, Wanrong Zhu, William Yang Wang, Xin Eric Wang","submitted_at":"2020-10-07T20:45:14Z","abstract_excerpt":"A major challenge in visually grounded language generation is to build robust benchmark datasets and models that can generalize well in real-world settings. To do this, it is critical to ensure that our evaluation protocols are correct, and benchmarks are reliable. In this work, we set forth to design a set of experiments to understand an important but often ignored problem in visually grounded language generation: given that humans have different utilities and visual attention, how will the sample variance in multi-reference datasets affect the models' performance? Empirically, we study sever"},"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":"2010.03644","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-10-07T20:45:14Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"6d4ef244486f1d0d55c841bb29e6e395f97b2fd53df1f3b4f36d9e28f61371d4","abstract_canon_sha256":"de71e4794cd8390bc3c552c9a50f5ed9042b49cd931a7efe48cb1e8fbbbc031e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:41:21.554076Z","signature_b64":"c5B/Sc/sEnCcySEpOOM8WJ52w/N//lGBqiQD5AsrGl1HJx9ApFIucZe9tzDMEV8H54okdF3t26I5zV7qVLnBDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a4d305dd67dc4aaa8360342c450f8b31817e108f6c780f6b53e888541e398391","last_reissued_at":"2026-07-05T01:41:21.553719Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:41:21.553719Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards Understanding Sample Variance in Visually Grounded Language Generation: Evaluations and Observations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.CL","authors_text":"Kazoo Sone, Pradyumna Narayana, Sugato Basu, Wanrong Zhu, William Yang Wang, Xin Eric Wang","submitted_at":"2020-10-07T20:45:14Z","abstract_excerpt":"A major challenge in visually grounded language generation is to build robust benchmark datasets and models that can generalize well in real-world settings. To do this, it is critical to ensure that our evaluation protocols are correct, and benchmarks are reliable. In this work, we set forth to design a set of experiments to understand an important but often ignored problem in visually grounded language generation: given that humans have different utilities and visual attention, how will the sample variance in multi-reference datasets affect the models' performance? Empirically, we study sever"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2010.03644","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/2010.03644/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":"2010.03644","created_at":"2026-07-05T01:41:21.553773+00:00"},{"alias_kind":"arxiv_version","alias_value":"2010.03644v1","created_at":"2026-07-05T01:41:21.553773+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2010.03644","created_at":"2026-07-05T01:41:21.553773+00:00"},{"alias_kind":"pith_short_12","alias_value":"UTJQLXLH3RFK","created_at":"2026-07-05T01:41:21.553773+00:00"},{"alias_kind":"pith_short_16","alias_value":"UTJQLXLH3RFKVA3A","created_at":"2026-07-05T01:41:21.553773+00:00"},{"alias_kind":"pith_short_8","alias_value":"UTJQLXLH","created_at":"2026-07-05T01:41:21.553773+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2411.14688","citing_title":"Whats in a Video: Factorized Autoregressive Decoding for Online Dense Video Captioning","ref_index":76,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UTJQLXLH3RFKVA3AGQWEKD4LGG","json":"https://pith.science/pith/UTJQLXLH3RFKVA3AGQWEKD4LGG.json","graph_json":"https://pith.science/api/pith-number/UTJQLXLH3RFKVA3AGQWEKD4LGG/graph.json","events_json":"https://pith.science/api/pith-number/UTJQLXLH3RFKVA3AGQWEKD4LGG/events.json","paper":"https://pith.science/paper/UTJQLXLH"},"agent_actions":{"view_html":"https://pith.science/pith/UTJQLXLH3RFKVA3AGQWEKD4LGG","download_json":"https://pith.science/pith/UTJQLXLH3RFKVA3AGQWEKD4LGG.json","view_paper":"https://pith.science/paper/UTJQLXLH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2010.03644&json=true","fetch_graph":"https://pith.science/api/pith-number/UTJQLXLH3RFKVA3AGQWEKD4LGG/graph.json","fetch_events":"https://pith.science/api/pith-number/UTJQLXLH3RFKVA3AGQWEKD4LGG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UTJQLXLH3RFKVA3AGQWEKD4LGG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UTJQLXLH3RFKVA3AGQWEKD4LGG/action/storage_attestation","attest_author":"https://pith.science/pith/UTJQLXLH3RFKVA3AGQWEKD4LGG/action/author_attestation","sign_citation":"https://pith.science/pith/UTJQLXLH3RFKVA3AGQWEKD4LGG/action/citation_signature","submit_replication":"https://pith.science/pith/UTJQLXLH3RFKVA3AGQWEKD4LGG/action/replication_record"}},"created_at":"2026-07-05T01:41:21.553773+00:00","updated_at":"2026-07-05T01:41:21.553773+00:00"}