{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:OQFTPECJ3FJR2QV3BIGOPZZCJW","short_pith_number":"pith:OQFTPECJ","schema_version":"1.0","canonical_sha256":"740b379049d9531d42bb0a0ce7e7224dbe61024be159a08bbf08a1739ee4d66e","source":{"kind":"arxiv","id":"2409.10702","version":2},"attestation_state":"computed","paper":{"title":"Model-in-the-Loop (MILO): Accelerating Multimodal AI Data Annotation with LLMs","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.HC","authors_text":"Bob Kamma, Boying Gong, Daniel Lee, David Stevens, Jiabo Hu, Miao Liu, Na Li, Ning Zhang, Pranay Shah, Robert Kuo, Wenwen Jiang, Xu Chen, Yifan Wang","submitted_at":"2024-09-16T20:05:57Z","abstract_excerpt":"The growing demand for AI training data has transformed data annotation into a global industry, but traditional approaches relying on human annotators are often time-consuming, labor-intensive, and prone to inconsistent quality. We propose the Model-in-the-Loop (MILO) framework, which integrates AI/ML models into the annotation process. Our research introduces a collaborative paradigm that leverages the strengths of both professional human annotators and large language models (LLMs). By employing LLMs as pre-annotation and real-time assistants, and judges on annotator responses, MILO enables e"},"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":"2409.10702","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.HC","submitted_at":"2024-09-16T20:05:57Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"e66c5a0405fc2d5d14df7d989e06c0164429c1709c98064cf58e28eca249e127","abstract_canon_sha256":"74ec52b1a1f0814adf14cd7115559709cbbeb829bed4aef6abebb2ef46061d5f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:10:50.520415Z","signature_b64":"s+Yc48KfnfH+Kf06F4XLMZc5XbOeM42syARFUQAwGMRrY0IeXaG/6guxanTYfMJ0aSECVY4sXucG99fs/OiyCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"740b379049d9531d42bb0a0ce7e7224dbe61024be159a08bbf08a1739ee4d66e","last_reissued_at":"2026-07-05T09:10:50.519923Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:10:50.519923Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Model-in-the-Loop (MILO): Accelerating Multimodal AI Data Annotation with LLMs","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.HC","authors_text":"Bob Kamma, Boying Gong, Daniel Lee, David Stevens, Jiabo Hu, Miao Liu, Na Li, Ning Zhang, Pranay Shah, Robert Kuo, Wenwen Jiang, Xu Chen, Yifan Wang","submitted_at":"2024-09-16T20:05:57Z","abstract_excerpt":"The growing demand for AI training data has transformed data annotation into a global industry, but traditional approaches relying on human annotators are often time-consuming, labor-intensive, and prone to inconsistent quality. We propose the Model-in-the-Loop (MILO) framework, which integrates AI/ML models into the annotation process. Our research introduces a collaborative paradigm that leverages the strengths of both professional human annotators and large language models (LLMs). By employing LLMs as pre-annotation and real-time assistants, and judges on annotator responses, MILO enables e"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.10702","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2409.10702/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":"2409.10702","created_at":"2026-07-05T09:10:50.519981+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.10702v2","created_at":"2026-07-05T09:10:50.519981+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.10702","created_at":"2026-07-05T09:10:50.519981+00:00"},{"alias_kind":"pith_short_12","alias_value":"OQFTPECJ3FJR","created_at":"2026-07-05T09:10:50.519981+00:00"},{"alias_kind":"pith_short_16","alias_value":"OQFTPECJ3FJR2QV3","created_at":"2026-07-05T09:10:50.519981+00:00"},{"alias_kind":"pith_short_8","alias_value":"OQFTPECJ","created_at":"2026-07-05T09:10:50.519981+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.02200","citing_title":"ARGUS: Policy-Adaptive Ad Governance via Evolving Reinforcement with Adversarial Umpiring","ref_index":95,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OQFTPECJ3FJR2QV3BIGOPZZCJW","json":"https://pith.science/pith/OQFTPECJ3FJR2QV3BIGOPZZCJW.json","graph_json":"https://pith.science/api/pith-number/OQFTPECJ3FJR2QV3BIGOPZZCJW/graph.json","events_json":"https://pith.science/api/pith-number/OQFTPECJ3FJR2QV3BIGOPZZCJW/events.json","paper":"https://pith.science/paper/OQFTPECJ"},"agent_actions":{"view_html":"https://pith.science/pith/OQFTPECJ3FJR2QV3BIGOPZZCJW","download_json":"https://pith.science/pith/OQFTPECJ3FJR2QV3BIGOPZZCJW.json","view_paper":"https://pith.science/paper/OQFTPECJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.10702&json=true","fetch_graph":"https://pith.science/api/pith-number/OQFTPECJ3FJR2QV3BIGOPZZCJW/graph.json","fetch_events":"https://pith.science/api/pith-number/OQFTPECJ3FJR2QV3BIGOPZZCJW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OQFTPECJ3FJR2QV3BIGOPZZCJW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OQFTPECJ3FJR2QV3BIGOPZZCJW/action/storage_attestation","attest_author":"https://pith.science/pith/OQFTPECJ3FJR2QV3BIGOPZZCJW/action/author_attestation","sign_citation":"https://pith.science/pith/OQFTPECJ3FJR2QV3BIGOPZZCJW/action/citation_signature","submit_replication":"https://pith.science/pith/OQFTPECJ3FJR2QV3BIGOPZZCJW/action/replication_record"}},"created_at":"2026-07-05T09:10:50.519981+00:00","updated_at":"2026-07-05T09:10:50.519981+00:00"}