{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:YESZ5ANBDBZXZDF4S3SETSETHE","short_pith_number":"pith:YESZ5ANB","schema_version":"1.0","canonical_sha256":"c1259e81a118737c8cbc96e449c893393d03edff41139dc06a85173786d6908a","source":{"kind":"arxiv","id":"2107.04313","version":1},"attestation_state":"computed","paper":{"title":"Memes in the Wild: Assessing the Generalizability of the Hateful Memes Challenge Dataset","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Aleksandar Shtedritski, Gal Wachtel, Hannah Rose Kirk, Martin Doff-Sotta, Noah Broestl, Paulius Rauba, Ruining Li, Xingjian Bai, Yennie Jun, Yuki M. Asano","submitted_at":"2021-07-09T09:04:05Z","abstract_excerpt":"Hateful memes pose a unique challenge for current machine learning systems because their message is derived from both text- and visual-modalities. To this effect, Facebook released the Hateful Memes Challenge, a dataset of memes with pre-extracted text captions, but it is unclear whether these synthetic examples generalize to `memes in the wild'. In this paper, we collect hateful and non-hateful memes from Pinterest to evaluate out-of-sample performance on models pre-trained on the Facebook dataset. We find that memes in the wild differ in two key aspects: 1) Captions must be extracted via OCR"},"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":"2107.04313","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-07-09T09:04:05Z","cross_cats_sorted":[],"title_canon_sha256":"0825e11552dda4da4010af38cf087450d189d87ad56e56af4a9ec2d4cf81844d","abstract_canon_sha256":"d2b686468a0b21f02801efe6387e2fd11e2ffc967694421eed1fb3d0b5313edc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:56:31.886044Z","signature_b64":"bLoM2WNTGSp7ZMrdqckzc16qXa+gHNFipN/C8HqW+3ugO8/XEhr/BBLGqC8GyT1ir8kiqs8g5SQ65MhGau9YBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c1259e81a118737c8cbc96e449c893393d03edff41139dc06a85173786d6908a","last_reissued_at":"2026-07-05T02:56:31.885536Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:56:31.885536Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Memes in the Wild: Assessing the Generalizability of the Hateful Memes Challenge Dataset","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Aleksandar Shtedritski, Gal Wachtel, Hannah Rose Kirk, Martin Doff-Sotta, Noah Broestl, Paulius Rauba, Ruining Li, Xingjian Bai, Yennie Jun, Yuki M. Asano","submitted_at":"2021-07-09T09:04:05Z","abstract_excerpt":"Hateful memes pose a unique challenge for current machine learning systems because their message is derived from both text- and visual-modalities. To this effect, Facebook released the Hateful Memes Challenge, a dataset of memes with pre-extracted text captions, but it is unclear whether these synthetic examples generalize to `memes in the wild'. In this paper, we collect hateful and non-hateful memes from Pinterest to evaluate out-of-sample performance on models pre-trained on the Facebook dataset. We find that memes in the wild differ in two key aspects: 1) Captions must be extracted via OCR"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.04313","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/2107.04313/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":"2107.04313","created_at":"2026-07-05T02:56:31.885593+00:00"},{"alias_kind":"arxiv_version","alias_value":"2107.04313v1","created_at":"2026-07-05T02:56:31.885593+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.04313","created_at":"2026-07-05T02:56:31.885593+00:00"},{"alias_kind":"pith_short_12","alias_value":"YESZ5ANBDBZX","created_at":"2026-07-05T02:56:31.885593+00:00"},{"alias_kind":"pith_short_16","alias_value":"YESZ5ANBDBZXZDF4","created_at":"2026-07-05T02:56:31.885593+00:00"},{"alias_kind":"pith_short_8","alias_value":"YESZ5ANB","created_at":"2026-07-05T02:56:31.885593+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.07086","citing_title":"Representation Decomposition for Learning Similarity and Contrastness Across Modalities for Affective Computing","ref_index":24,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YESZ5ANBDBZXZDF4S3SETSETHE","json":"https://pith.science/pith/YESZ5ANBDBZXZDF4S3SETSETHE.json","graph_json":"https://pith.science/api/pith-number/YESZ5ANBDBZXZDF4S3SETSETHE/graph.json","events_json":"https://pith.science/api/pith-number/YESZ5ANBDBZXZDF4S3SETSETHE/events.json","paper":"https://pith.science/paper/YESZ5ANB"},"agent_actions":{"view_html":"https://pith.science/pith/YESZ5ANBDBZXZDF4S3SETSETHE","download_json":"https://pith.science/pith/YESZ5ANBDBZXZDF4S3SETSETHE.json","view_paper":"https://pith.science/paper/YESZ5ANB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2107.04313&json=true","fetch_graph":"https://pith.science/api/pith-number/YESZ5ANBDBZXZDF4S3SETSETHE/graph.json","fetch_events":"https://pith.science/api/pith-number/YESZ5ANBDBZXZDF4S3SETSETHE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YESZ5ANBDBZXZDF4S3SETSETHE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YESZ5ANBDBZXZDF4S3SETSETHE/action/storage_attestation","attest_author":"https://pith.science/pith/YESZ5ANBDBZXZDF4S3SETSETHE/action/author_attestation","sign_citation":"https://pith.science/pith/YESZ5ANBDBZXZDF4S3SETSETHE/action/citation_signature","submit_replication":"https://pith.science/pith/YESZ5ANBDBZXZDF4S3SETSETHE/action/replication_record"}},"created_at":"2026-07-05T02:56:31.885593+00:00","updated_at":"2026-07-05T02:56:31.885593+00:00"}