{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:HSZ6BQMWZBC3XSKNPJTEMVYSSK","short_pith_number":"pith:HSZ6BQMW","schema_version":"1.0","canonical_sha256":"3cb3e0c196c845bbc94d7a6646571292af29083cfbb592cc6b8941bef4fe9b42","source":{"kind":"arxiv","id":"2101.06278","version":3},"attestation_state":"computed","paper":{"title":"COSMOS: Catching Out-of-Context Misinformation with Self-Supervised Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Chris Bregler, Matthias Nie{\\ss}ner, Shivangi Aneja","submitted_at":"2021-01-15T19:00:42Z","abstract_excerpt":"Despite the recent attention to DeepFakes, one of the most prevalent ways to mislead audiences on social media is the use of unaltered images in a new but false context. To address these challenges and support fact-checkers, we propose a new method that automatically detects out-of-context image and text pairs. Our key insight is to leverage the grounding of image with text to distinguish out-of-context scenarios that cannot be disambiguated with language alone. We propose a self-supervised training strategy where we only need a set of captioned images. At train time, our method learns to sele"},"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":"2101.06278","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-01-15T19:00:42Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"a9e285c354051d9d8d046fadbce1b6f81ce5c0d69705b2bb8e7978fb8b3eec14","abstract_canon_sha256":"3a9a716dea94ac4de2b95fdb17796b27af46eaa31364e8639d9a566d97b301eb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:34:10.961817Z","signature_b64":"pCjbgUGFws0GU/hx1MZ837MgBKEhT7dUHER+2iiMXBQBb5g1/wfqF9SisHm7HNzBBhvQFseODhF1NfVOinfWDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3cb3e0c196c845bbc94d7a6646571292af29083cfbb592cc6b8941bef4fe9b42","last_reissued_at":"2026-07-05T02:34:10.961319Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:34:10.961319Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"COSMOS: Catching Out-of-Context Misinformation with Self-Supervised Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Chris Bregler, Matthias Nie{\\ss}ner, Shivangi Aneja","submitted_at":"2021-01-15T19:00:42Z","abstract_excerpt":"Despite the recent attention to DeepFakes, one of the most prevalent ways to mislead audiences on social media is the use of unaltered images in a new but false context. To address these challenges and support fact-checkers, we propose a new method that automatically detects out-of-context image and text pairs. Our key insight is to leverage the grounding of image with text to distinguish out-of-context scenarios that cannot be disambiguated with language alone. We propose a self-supervised training strategy where we only need a set of captioned images. At train time, our method learns to sele"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2101.06278","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/2101.06278/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":"2101.06278","created_at":"2026-07-05T02:34:10.961387+00:00"},{"alias_kind":"arxiv_version","alias_value":"2101.06278v3","created_at":"2026-07-05T02:34:10.961387+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2101.06278","created_at":"2026-07-05T02:34:10.961387+00:00"},{"alias_kind":"pith_short_12","alias_value":"HSZ6BQMWZBC3","created_at":"2026-07-05T02:34:10.961387+00:00"},{"alias_kind":"pith_short_16","alias_value":"HSZ6BQMWZBC3XSKN","created_at":"2026-07-05T02:34:10.961387+00:00"},{"alias_kind":"pith_short_8","alias_value":"HSZ6BQMW","created_at":"2026-07-05T02:34:10.961387+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24627","citing_title":"The Warrant Gap: Claim-Conditioned Re-scoring for Fact-Checking","ref_index":90,"is_internal_anchor":false},{"citing_arxiv_id":"2606.24112","citing_title":"ReMMD: Realistic Multilingual Multi-Image Agentic Verification for Multimodal Misinformation Detection","ref_index":54,"is_internal_anchor":false},{"citing_arxiv_id":"2606.22339","citing_title":"T-IMPACT: A Severity-Aware Benchmark for Contextual Image-Text Manipulation","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2606.04098","citing_title":"When Seeing Is Not Believing -- A Benchmark for Search-Grounded Video Misinformation Detection","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2606.03348","citing_title":"SynCred-Bench: Benchmarking Synthetic Credibility in AI-Generated Visual Misinformation","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2504.10166","citing_title":"Fact-Checking with Contextual Narratives: Leveraging Retrieval-Augmented LLMs for Social Media Analysis","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2601.08611","citing_title":"VeriTaS: The First Dynamic Benchmark for Multimodal Automated Fact-Checking","ref_index":1,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HSZ6BQMWZBC3XSKNPJTEMVYSSK","json":"https://pith.science/pith/HSZ6BQMWZBC3XSKNPJTEMVYSSK.json","graph_json":"https://pith.science/api/pith-number/HSZ6BQMWZBC3XSKNPJTEMVYSSK/graph.json","events_json":"https://pith.science/api/pith-number/HSZ6BQMWZBC3XSKNPJTEMVYSSK/events.json","paper":"https://pith.science/paper/HSZ6BQMW"},"agent_actions":{"view_html":"https://pith.science/pith/HSZ6BQMWZBC3XSKNPJTEMVYSSK","download_json":"https://pith.science/pith/HSZ6BQMWZBC3XSKNPJTEMVYSSK.json","view_paper":"https://pith.science/paper/HSZ6BQMW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2101.06278&json=true","fetch_graph":"https://pith.science/api/pith-number/HSZ6BQMWZBC3XSKNPJTEMVYSSK/graph.json","fetch_events":"https://pith.science/api/pith-number/HSZ6BQMWZBC3XSKNPJTEMVYSSK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HSZ6BQMWZBC3XSKNPJTEMVYSSK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HSZ6BQMWZBC3XSKNPJTEMVYSSK/action/storage_attestation","attest_author":"https://pith.science/pith/HSZ6BQMWZBC3XSKNPJTEMVYSSK/action/author_attestation","sign_citation":"https://pith.science/pith/HSZ6BQMWZBC3XSKNPJTEMVYSSK/action/citation_signature","submit_replication":"https://pith.science/pith/HSZ6BQMWZBC3XSKNPJTEMVYSSK/action/replication_record"}},"created_at":"2026-07-05T02:34:10.961387+00:00","updated_at":"2026-07-05T02:34:10.961387+00:00"}