{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:QBQRWSWITBVP2GIRBFKOBBSQ2T","short_pith_number":"pith:QBQRWSWI","schema_version":"1.0","canonical_sha256":"80611b4ac8986afd19110954e08650d4db37e6b38816e4eaebb9ed82b5824209","source":{"kind":"arxiv","id":"2408.10086","version":1},"attestation_state":"computed","paper":{"title":"ARMADA: Attribute-Based Multimodal Data Augmentation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Heng Ji, Jeonghwan Kim, Kuan-Hao Huang, Nanyun Peng, Te-Lin Wu, Xiaomeng Jin, Yu Zhou","submitted_at":"2024-08-19T15:27:25Z","abstract_excerpt":"In Multimodal Language Models (MLMs), the cost of manually annotating high-quality image-text pair data for fine-tuning and alignment is extremely high. While existing multimodal data augmentation frameworks propose ways to augment image-text pairs, they either suffer from semantic inconsistency between texts and images, or generate unrealistic images, causing knowledge gap with real world examples. To address these issues, we propose Attribute-based Multimodal Data Augmentation (ARMADA), a novel multimodal data augmentation method via knowledge-guided manipulation of visual attributes of the "},"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":"2408.10086","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-08-19T15:27:25Z","cross_cats_sorted":[],"title_canon_sha256":"809cc0d014893993fbcd75a1b7e3688fccb8a62aba391cd38025a0d8fc573c04","abstract_canon_sha256":"6b0830ace6ad5dc4280a26bad0c4baa48c7fbb36f964b04d725903159820ebdc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:56:51.131982Z","signature_b64":"tfLMCV7lj9hZn7a8cN6FyX1T9GBYHGoU8v+xtPY+ZN73B14YHX7u+iUk/HfiMavrLXeBfo8+7bJz19gXWf2YDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"80611b4ac8986afd19110954e08650d4db37e6b38816e4eaebb9ed82b5824209","last_reissued_at":"2026-07-05T08:56:51.131559Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:56:51.131559Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ARMADA: Attribute-Based Multimodal Data Augmentation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Heng Ji, Jeonghwan Kim, Kuan-Hao Huang, Nanyun Peng, Te-Lin Wu, Xiaomeng Jin, Yu Zhou","submitted_at":"2024-08-19T15:27:25Z","abstract_excerpt":"In Multimodal Language Models (MLMs), the cost of manually annotating high-quality image-text pair data for fine-tuning and alignment is extremely high. While existing multimodal data augmentation frameworks propose ways to augment image-text pairs, they either suffer from semantic inconsistency between texts and images, or generate unrealistic images, causing knowledge gap with real world examples. To address these issues, we propose Attribute-based Multimodal Data Augmentation (ARMADA), a novel multimodal data augmentation method via knowledge-guided manipulation of visual attributes of the "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.10086","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/2408.10086/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":"2408.10086","created_at":"2026-07-05T08:56:51.131620+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.10086v1","created_at":"2026-07-05T08:56:51.131620+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.10086","created_at":"2026-07-05T08:56:51.131620+00:00"},{"alias_kind":"pith_short_12","alias_value":"QBQRWSWITBVP","created_at":"2026-07-05T08:56:51.131620+00:00"},{"alias_kind":"pith_short_16","alias_value":"QBQRWSWITBVP2GIR","created_at":"2026-07-05T08:56:51.131620+00:00"},{"alias_kind":"pith_short_8","alias_value":"QBQRWSWI","created_at":"2026-07-05T08:56:51.131620+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.17695","citing_title":"SynRES: Towards Referring Expression Segmentation in the Wild via Synthetic Data","ref_index":14,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QBQRWSWITBVP2GIRBFKOBBSQ2T","json":"https://pith.science/pith/QBQRWSWITBVP2GIRBFKOBBSQ2T.json","graph_json":"https://pith.science/api/pith-number/QBQRWSWITBVP2GIRBFKOBBSQ2T/graph.json","events_json":"https://pith.science/api/pith-number/QBQRWSWITBVP2GIRBFKOBBSQ2T/events.json","paper":"https://pith.science/paper/QBQRWSWI"},"agent_actions":{"view_html":"https://pith.science/pith/QBQRWSWITBVP2GIRBFKOBBSQ2T","download_json":"https://pith.science/pith/QBQRWSWITBVP2GIRBFKOBBSQ2T.json","view_paper":"https://pith.science/paper/QBQRWSWI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.10086&json=true","fetch_graph":"https://pith.science/api/pith-number/QBQRWSWITBVP2GIRBFKOBBSQ2T/graph.json","fetch_events":"https://pith.science/api/pith-number/QBQRWSWITBVP2GIRBFKOBBSQ2T/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QBQRWSWITBVP2GIRBFKOBBSQ2T/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QBQRWSWITBVP2GIRBFKOBBSQ2T/action/storage_attestation","attest_author":"https://pith.science/pith/QBQRWSWITBVP2GIRBFKOBBSQ2T/action/author_attestation","sign_citation":"https://pith.science/pith/QBQRWSWITBVP2GIRBFKOBBSQ2T/action/citation_signature","submit_replication":"https://pith.science/pith/QBQRWSWITBVP2GIRBFKOBBSQ2T/action/replication_record"}},"created_at":"2026-07-05T08:56:51.131620+00:00","updated_at":"2026-07-05T08:56:51.131620+00:00"}