{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:E3BHYDTPBYZDTNPBZQMUHE27TS","short_pith_number":"pith:E3BHYDTP","schema_version":"1.0","canonical_sha256":"26c27c0e6f0e3239b5e1cc1943935f9cb202d2efc7387316cc2e3c65b8b30d54","source":{"kind":"arxiv","id":"2104.07511","version":3},"attestation_state":"computed","paper":{"title":"Ensemble of MRR and NDCG models for Visual Dialog","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.CV","cs.IR","cs.LG"],"primary_cat":"cs.AI","authors_text":"Idan Schwartz","submitted_at":"2021-04-15T15:09:32Z","abstract_excerpt":"Assessing an AI agent that can converse in human language and understand visual content is challenging. Generation metrics, such as BLEU scores favor correct syntax over semantics. Hence a discriminative approach is often used, where an agent ranks a set of candidate options. The mean reciprocal rank (MRR) metric evaluates the model performance by taking into account the rank of a single human-derived answer. This approach, however, raises a new challenge: the ambiguity and synonymy of answers, for instance, semantic equivalence (e.g., `yeah' and `yes'). To address this, the normalized discoun"},"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":"2104.07511","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2021-04-15T15:09:32Z","cross_cats_sorted":["cs.CL","cs.CV","cs.IR","cs.LG"],"title_canon_sha256":"e605eb4afcd3d8e27a6737e5db2cb87e04ccdaa25a117283ce88392c5f2f761c","abstract_canon_sha256":"8ddd1fb8b23fb2b2d916fa971a946f15561b419c1c747158cf3957e6b3a15e97"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:03:19.529115Z","signature_b64":"1MnKg2L/3FimQS4DQADD1xeEEETuf9onglIy8LCApgC8tiP0eMQ3LDr2SpQsQu51EY0YDFnWJtrG/rj2M07+BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"26c27c0e6f0e3239b5e1cc1943935f9cb202d2efc7387316cc2e3c65b8b30d54","last_reissued_at":"2026-07-05T03:03:19.528615Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:03:19.528615Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Ensemble of MRR and NDCG models for Visual Dialog","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.CV","cs.IR","cs.LG"],"primary_cat":"cs.AI","authors_text":"Idan Schwartz","submitted_at":"2021-04-15T15:09:32Z","abstract_excerpt":"Assessing an AI agent that can converse in human language and understand visual content is challenging. Generation metrics, such as BLEU scores favor correct syntax over semantics. Hence a discriminative approach is often used, where an agent ranks a set of candidate options. The mean reciprocal rank (MRR) metric evaluates the model performance by taking into account the rank of a single human-derived answer. This approach, however, raises a new challenge: the ambiguity and synonymy of answers, for instance, semantic equivalence (e.g., `yeah' and `yes'). To address this, the normalized discoun"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2104.07511","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/2104.07511/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":"2104.07511","created_at":"2026-07-05T03:03:19.528681+00:00"},{"alias_kind":"arxiv_version","alias_value":"2104.07511v3","created_at":"2026-07-05T03:03:19.528681+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2104.07511","created_at":"2026-07-05T03:03:19.528681+00:00"},{"alias_kind":"pith_short_12","alias_value":"E3BHYDTPBYZD","created_at":"2026-07-05T03:03:19.528681+00:00"},{"alias_kind":"pith_short_16","alias_value":"E3BHYDTPBYZDTNPB","created_at":"2026-07-05T03:03:19.528681+00:00"},{"alias_kind":"pith_short_8","alias_value":"E3BHYDTP","created_at":"2026-07-05T03:03:19.528681+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.12425","citing_title":"Advancing Retrieval-Augmented Generation for Structured Enterprise and Internal Data","ref_index":12,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/E3BHYDTPBYZDTNPBZQMUHE27TS","json":"https://pith.science/pith/E3BHYDTPBYZDTNPBZQMUHE27TS.json","graph_json":"https://pith.science/api/pith-number/E3BHYDTPBYZDTNPBZQMUHE27TS/graph.json","events_json":"https://pith.science/api/pith-number/E3BHYDTPBYZDTNPBZQMUHE27TS/events.json","paper":"https://pith.science/paper/E3BHYDTP"},"agent_actions":{"view_html":"https://pith.science/pith/E3BHYDTPBYZDTNPBZQMUHE27TS","download_json":"https://pith.science/pith/E3BHYDTPBYZDTNPBZQMUHE27TS.json","view_paper":"https://pith.science/paper/E3BHYDTP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2104.07511&json=true","fetch_graph":"https://pith.science/api/pith-number/E3BHYDTPBYZDTNPBZQMUHE27TS/graph.json","fetch_events":"https://pith.science/api/pith-number/E3BHYDTPBYZDTNPBZQMUHE27TS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/E3BHYDTPBYZDTNPBZQMUHE27TS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/E3BHYDTPBYZDTNPBZQMUHE27TS/action/storage_attestation","attest_author":"https://pith.science/pith/E3BHYDTPBYZDTNPBZQMUHE27TS/action/author_attestation","sign_citation":"https://pith.science/pith/E3BHYDTPBYZDTNPBZQMUHE27TS/action/citation_signature","submit_replication":"https://pith.science/pith/E3BHYDTPBYZDTNPBZQMUHE27TS/action/replication_record"}},"created_at":"2026-07-05T03:03:19.528681+00:00","updated_at":"2026-07-05T03:03:19.528681+00:00"}