{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:B4MIHDONV3VNLWX6KRYQGJWYUD","short_pith_number":"pith:B4MIHDON","schema_version":"1.0","canonical_sha256":"0f18838dcdaeead5dafe54710326d8a0f76103f1fe794b195768e67aa191850a","source":{"kind":"arxiv","id":"2405.14030","version":3},"attestation_state":"computed","paper":{"title":"Refining Skewed Perceptions in Vision-Language Contrastive Models through Visual Representations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Haocheng Dai, Sarang Joshi","submitted_at":"2024-05-22T22:03:11Z","abstract_excerpt":"Large vision-language contrastive models (VLCMs), such as CLIP, have become foundational, demonstrating remarkable success across a variety of downstream tasks. Despite their advantages, these models, akin to other foundational systems, inherit biases from the disproportionate distribution of real-world data, leading to misconceptions about the actual environment. Prevalent datasets like ImageNet are often riddled with non-causal, spurious correlations that can diminish VLCM performance in scenarios where these contextual elements are absent. This study presents an investigation into how a sim"},"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":"2405.14030","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-05-22T22:03:11Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"5a95326665e5c85dec720b287a8fd85574271a278815fffed3743c96f136d017","abstract_canon_sha256":"ef758c5189433993377b27df6f68ba69d2a26ad20af3f077e6650753c92cff46"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:33:58.240979Z","signature_b64":"xI88AD9nT/QfzoL3JV1L4WrG6GU8cIXGr9OqJh7vxIKkUCLDuNe8Vt3FQoAjc8PfIbQeH3nSK/ZUYVO0KI5CAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0f18838dcdaeead5dafe54710326d8a0f76103f1fe794b195768e67aa191850a","last_reissued_at":"2026-07-05T11:33:58.240482Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:33:58.240482Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Refining Skewed Perceptions in Vision-Language Contrastive Models through Visual Representations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Haocheng Dai, Sarang Joshi","submitted_at":"2024-05-22T22:03:11Z","abstract_excerpt":"Large vision-language contrastive models (VLCMs), such as CLIP, have become foundational, demonstrating remarkable success across a variety of downstream tasks. Despite their advantages, these models, akin to other foundational systems, inherit biases from the disproportionate distribution of real-world data, leading to misconceptions about the actual environment. Prevalent datasets like ImageNet are often riddled with non-causal, spurious correlations that can diminish VLCM performance in scenarios where these contextual elements are absent. This study presents an investigation into how a sim"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.14030","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/2405.14030/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":"2405.14030","created_at":"2026-07-05T11:33:58.240540+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.14030v3","created_at":"2026-07-05T11:33:58.240540+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.14030","created_at":"2026-07-05T11:33:58.240540+00:00"},{"alias_kind":"pith_short_12","alias_value":"B4MIHDONV3VN","created_at":"2026-07-05T11:33:58.240540+00:00"},{"alias_kind":"pith_short_16","alias_value":"B4MIHDONV3VNLWX6","created_at":"2026-07-05T11:33:58.240540+00:00"},{"alias_kind":"pith_short_8","alias_value":"B4MIHDON","created_at":"2026-07-05T11:33:58.240540+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.08979","citing_title":"PRISM: Reducing Spurious Implicit Biases in Vision-Language Models with LLM-Guided Embedding Projection","ref_index":6,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/B4MIHDONV3VNLWX6KRYQGJWYUD","json":"https://pith.science/pith/B4MIHDONV3VNLWX6KRYQGJWYUD.json","graph_json":"https://pith.science/api/pith-number/B4MIHDONV3VNLWX6KRYQGJWYUD/graph.json","events_json":"https://pith.science/api/pith-number/B4MIHDONV3VNLWX6KRYQGJWYUD/events.json","paper":"https://pith.science/paper/B4MIHDON"},"agent_actions":{"view_html":"https://pith.science/pith/B4MIHDONV3VNLWX6KRYQGJWYUD","download_json":"https://pith.science/pith/B4MIHDONV3VNLWX6KRYQGJWYUD.json","view_paper":"https://pith.science/paper/B4MIHDON","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.14030&json=true","fetch_graph":"https://pith.science/api/pith-number/B4MIHDONV3VNLWX6KRYQGJWYUD/graph.json","fetch_events":"https://pith.science/api/pith-number/B4MIHDONV3VNLWX6KRYQGJWYUD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/B4MIHDONV3VNLWX6KRYQGJWYUD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/B4MIHDONV3VNLWX6KRYQGJWYUD/action/storage_attestation","attest_author":"https://pith.science/pith/B4MIHDONV3VNLWX6KRYQGJWYUD/action/author_attestation","sign_citation":"https://pith.science/pith/B4MIHDONV3VNLWX6KRYQGJWYUD/action/citation_signature","submit_replication":"https://pith.science/pith/B4MIHDONV3VNLWX6KRYQGJWYUD/action/replication_record"}},"created_at":"2026-07-05T11:33:58.240540+00:00","updated_at":"2026-07-05T11:33:58.240540+00:00"}