{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:Z5ZFZT3UWTOUOBUS2ZHTQUYCIE","short_pith_number":"pith:Z5ZFZT3U","schema_version":"1.0","canonical_sha256":"cf725ccf74b4dd470692d64f385302413718cd628760489798cccacb9c9c4f7a","source":{"kind":"arxiv","id":"2211.06516","version":1},"attestation_state":"computed","paper":{"title":"Bandits for Online Calibration: An Application to Content Moderation on Social Media Platforms","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Caner Gocmen, Christopher Palow, Daniel Haimovich, Darren Hwang, Deeksha Sinha, Dima Karamshuk, Gregory Macnamara, Hamsa Bastani, Jake Mullett, Jiayuan Ma, Kevin Schaeffer, Nicolas Stier-Moses, Omar Abdul Baki, Osbert Bastani, Parikshit Shah, Peng Xu, Sung Park, Thomas Leeper, Varun S Rajagopal, Vashist Avadhanula","submitted_at":"2022-11-11T23:55:53Z","abstract_excerpt":"We describe the current content moderation strategy employed by Meta to remove policy-violating content from its platforms. Meta relies on both handcrafted and learned risk models to flag potentially violating content for human review. Our approach aggregates these risk models into a single ranking score, calibrating them to prioritize more reliable risk models. A key challenge is that violation trends change over time, affecting which risk models are most reliable. Our system additionally handles production challenges such as changing risk models and novel risk models. We use a contextual ban"},"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":"2211.06516","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2022-11-11T23:55:53Z","cross_cats_sorted":[],"title_canon_sha256":"1a3a6070708c8b93d9e1b45f64a7ba0b5f5fc79f8e1f261fa85e38c719947145","abstract_canon_sha256":"cd345c2b14cc7ada86a753e05cc2782e11377275e4c43838f94705ad6f8b8ee9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:15:25.903202Z","signature_b64":"yrxbIKOTPHjl8FPwIpGWAt89MdI74k0wx4fNxoSLHZqJLtMWjGViy27qMzle7/w2zrYGZK19/HuGbcLF4poUCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cf725ccf74b4dd470692d64f385302413718cd628760489798cccacb9c9c4f7a","last_reissued_at":"2026-07-05T05:15:25.902703Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:15:25.902703Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Bandits for Online Calibration: An Application to Content Moderation on Social Media Platforms","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Caner Gocmen, Christopher Palow, Daniel Haimovich, Darren Hwang, Deeksha Sinha, Dima Karamshuk, Gregory Macnamara, Hamsa Bastani, Jake Mullett, Jiayuan Ma, Kevin Schaeffer, Nicolas Stier-Moses, Omar Abdul Baki, Osbert Bastani, Parikshit Shah, Peng Xu, Sung Park, Thomas Leeper, Varun S Rajagopal, Vashist Avadhanula","submitted_at":"2022-11-11T23:55:53Z","abstract_excerpt":"We describe the current content moderation strategy employed by Meta to remove policy-violating content from its platforms. Meta relies on both handcrafted and learned risk models to flag potentially violating content for human review. Our approach aggregates these risk models into a single ranking score, calibrating them to prioritize more reliable risk models. A key challenge is that violation trends change over time, affecting which risk models are most reliable. Our system additionally handles production challenges such as changing risk models and novel risk models. We use a contextual ban"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2211.06516","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/2211.06516/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":"2211.06516","created_at":"2026-07-05T05:15:25.902771+00:00"},{"alias_kind":"arxiv_version","alias_value":"2211.06516v1","created_at":"2026-07-05T05:15:25.902771+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2211.06516","created_at":"2026-07-05T05:15:25.902771+00:00"},{"alias_kind":"pith_short_12","alias_value":"Z5ZFZT3UWTOU","created_at":"2026-07-05T05:15:25.902771+00:00"},{"alias_kind":"pith_short_16","alias_value":"Z5ZFZT3UWTOUOBUS","created_at":"2026-07-05T05:15:25.902771+00:00"},{"alias_kind":"pith_short_8","alias_value":"Z5ZFZT3U","created_at":"2026-07-05T05:15:25.902771+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.09075","citing_title":"Optimality of Sub-network Laplace Approximations: New Results and Methods","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12289","citing_title":"The Enforcement and Feasibility of Hate Speech Moderation on Twitter","ref_index":45,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Z5ZFZT3UWTOUOBUS2ZHTQUYCIE","json":"https://pith.science/pith/Z5ZFZT3UWTOUOBUS2ZHTQUYCIE.json","graph_json":"https://pith.science/api/pith-number/Z5ZFZT3UWTOUOBUS2ZHTQUYCIE/graph.json","events_json":"https://pith.science/api/pith-number/Z5ZFZT3UWTOUOBUS2ZHTQUYCIE/events.json","paper":"https://pith.science/paper/Z5ZFZT3U"},"agent_actions":{"view_html":"https://pith.science/pith/Z5ZFZT3UWTOUOBUS2ZHTQUYCIE","download_json":"https://pith.science/pith/Z5ZFZT3UWTOUOBUS2ZHTQUYCIE.json","view_paper":"https://pith.science/paper/Z5ZFZT3U","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2211.06516&json=true","fetch_graph":"https://pith.science/api/pith-number/Z5ZFZT3UWTOUOBUS2ZHTQUYCIE/graph.json","fetch_events":"https://pith.science/api/pith-number/Z5ZFZT3UWTOUOBUS2ZHTQUYCIE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Z5ZFZT3UWTOUOBUS2ZHTQUYCIE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Z5ZFZT3UWTOUOBUS2ZHTQUYCIE/action/storage_attestation","attest_author":"https://pith.science/pith/Z5ZFZT3UWTOUOBUS2ZHTQUYCIE/action/author_attestation","sign_citation":"https://pith.science/pith/Z5ZFZT3UWTOUOBUS2ZHTQUYCIE/action/citation_signature","submit_replication":"https://pith.science/pith/Z5ZFZT3UWTOUOBUS2ZHTQUYCIE/action/replication_record"}},"created_at":"2026-07-05T05:15:25.902771+00:00","updated_at":"2026-07-05T05:15:25.902771+00:00"}