{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:WXRB6AGYI4PDJI5IFZAS5XFBDA","short_pith_number":"pith:WXRB6AGY","schema_version":"1.0","canonical_sha256":"b5e21f00d8471e34a3a82e412edca1180a042fc369a3c3d94bd8e96c7efa61eb","source":{"kind":"arxiv","id":"1712.00377","version":2},"attestation_state":"computed","paper":{"title":"Don't Just Assume; Look and Answer: Overcoming Priors for Visual Question Answering","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Aishwarya Agrawal, Aniruddha Kembhavi, Devi Parikh, Dhruv Batra","submitted_at":"2017-12-01T15:48:50Z","abstract_excerpt":"A number of studies have found that today's Visual Question Answering (VQA) models are heavily driven by superficial correlations in the training data and lack sufficient image grounding. To encourage development of models geared towards the latter, we propose a new setting for VQA where for every question type, train and test sets have different prior distributions of answers. Specifically, we present new splits of the VQA v1 and VQA v2 datasets, which we call Visual Question Answering under Changing Priors (VQA-CP v1 and VQA-CP v2 respectively). First, we evaluate several existing VQA models"},"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":"1712.00377","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2017-12-01T15:48:50Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"28f5bd89100317e874a25e460e8906314114652e6b1bd8f765d72c52db93ad93","abstract_canon_sha256":"8ec3e5615646e7171010f7b50d64d61c7a273c226496a6b2f62ac9d4f20b3b9b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:14:22.101599Z","signature_b64":"Agz8V28txlDx/hNdoQ3+d2dQ7H4wLV4qVmb/hHXaDhXPAihwPCv4Y3zZbUVbbrwQAUAqgVOWFmRoitP+zfeYDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b5e21f00d8471e34a3a82e412edca1180a042fc369a3c3d94bd8e96c7efa61eb","last_reissued_at":"2026-05-18T00:14:22.100800Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:14:22.100800Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Don't Just Assume; Look and Answer: Overcoming Priors for Visual Question Answering","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Aishwarya Agrawal, Aniruddha Kembhavi, Devi Parikh, Dhruv Batra","submitted_at":"2017-12-01T15:48:50Z","abstract_excerpt":"A number of studies have found that today's Visual Question Answering (VQA) models are heavily driven by superficial correlations in the training data and lack sufficient image grounding. To encourage development of models geared towards the latter, we propose a new setting for VQA where for every question type, train and test sets have different prior distributions of answers. Specifically, we present new splits of the VQA v1 and VQA v2 datasets, which we call Visual Question Answering under Changing Priors (VQA-CP v1 and VQA-CP v2 respectively). First, we evaluate several existing VQA models"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1712.00377","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":"1712.00377","created_at":"2026-05-18T00:14:22.100917+00:00"},{"alias_kind":"arxiv_version","alias_value":"1712.00377v2","created_at":"2026-05-18T00:14:22.100917+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1712.00377","created_at":"2026-05-18T00:14:22.100917+00:00"},{"alias_kind":"pith_short_12","alias_value":"WXRB6AGYI4PD","created_at":"2026-05-18T12:31:53.515858+00:00"},{"alias_kind":"pith_short_16","alias_value":"WXRB6AGYI4PDJI5I","created_at":"2026-05-18T12:31:53.515858+00:00"},{"alias_kind":"pith_short_8","alias_value":"WXRB6AGY","created_at":"2026-05-18T12:31:53.515858+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2308.08089","citing_title":"DragNUWA: Fine-grained Control in Video Generation by Integrating Text, Image, and Trajectory","ref_index":46,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WXRB6AGYI4PDJI5IFZAS5XFBDA","json":"https://pith.science/pith/WXRB6AGYI4PDJI5IFZAS5XFBDA.json","graph_json":"https://pith.science/api/pith-number/WXRB6AGYI4PDJI5IFZAS5XFBDA/graph.json","events_json":"https://pith.science/api/pith-number/WXRB6AGYI4PDJI5IFZAS5XFBDA/events.json","paper":"https://pith.science/paper/WXRB6AGY"},"agent_actions":{"view_html":"https://pith.science/pith/WXRB6AGYI4PDJI5IFZAS5XFBDA","download_json":"https://pith.science/pith/WXRB6AGYI4PDJI5IFZAS5XFBDA.json","view_paper":"https://pith.science/paper/WXRB6AGY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1712.00377&json=true","fetch_graph":"https://pith.science/api/pith-number/WXRB6AGYI4PDJI5IFZAS5XFBDA/graph.json","fetch_events":"https://pith.science/api/pith-number/WXRB6AGYI4PDJI5IFZAS5XFBDA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WXRB6AGYI4PDJI5IFZAS5XFBDA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WXRB6AGYI4PDJI5IFZAS5XFBDA/action/storage_attestation","attest_author":"https://pith.science/pith/WXRB6AGYI4PDJI5IFZAS5XFBDA/action/author_attestation","sign_citation":"https://pith.science/pith/WXRB6AGYI4PDJI5IFZAS5XFBDA/action/citation_signature","submit_replication":"https://pith.science/pith/WXRB6AGYI4PDJI5IFZAS5XFBDA/action/replication_record"}},"created_at":"2026-05-18T00:14:22.100917+00:00","updated_at":"2026-05-18T00:14:22.100917+00:00"}