{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:QOBZJVEDG772IENWJ23EMS4FUP","short_pith_number":"pith:QOBZJVED","schema_version":"1.0","canonical_sha256":"838394d48337ffa411b64eb6464b85a3e52c6c52677289e9fe1bb6cb2b569de4","source":{"kind":"arxiv","id":"2409.03277","version":3},"attestation_state":"computed","paper":{"title":"ChartMoE: Mixture of Diversely Aligned Expert Connector for Chart Understanding","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.CV"],"primary_cat":"cs.AI","authors_text":"Bowen Qu, Chengjin Xu, Chun Yuan, Jian Guo, Sinan Du, Yiyan Qi, Zhengzhuo Xu","submitted_at":"2024-09-05T06:41:02Z","abstract_excerpt":"Automatic chart understanding is crucial for content comprehension and document parsing. Multimodal Large Language Models (MLLMs) have demonstrated remarkable capabilities in chart understanding through domain-specific alignment and fine-tuning. However, current MLLMs still struggle to provide faithful data and reliable analysis only based on charts. To address it, we propose ChartMoE, which employs the Mixture of Expert (MoE) architecture to replace the traditional linear projector to bridge the modality gap. Specifically, we train several linear connectors through distinct alignment tasks, w"},"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":"2409.03277","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-09-05T06:41:02Z","cross_cats_sorted":["cs.CL","cs.CV"],"title_canon_sha256":"37545333aefd5eaeeecee4cd40611f30b12208558b7a3ae06593ad3e97f2bf17","abstract_canon_sha256":"543c93ce8bb12f08c38fd1b489c29e93a3d4811fbc57515ffc99ad46a30603c5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:30:56.326127Z","signature_b64":"PIxRu51nG/wdOi0gDBdRCSEoHY7pDrhgsx/ViOq3ATcbx7oZMzEAon1w797213D/7RHlHn0gd9YEO9Mb0u4jCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"838394d48337ffa411b64eb6464b85a3e52c6c52677289e9fe1bb6cb2b569de4","last_reissued_at":"2026-07-05T10:30:56.325385Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:30:56.325385Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ChartMoE: Mixture of Diversely Aligned Expert Connector for Chart Understanding","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.CV"],"primary_cat":"cs.AI","authors_text":"Bowen Qu, Chengjin Xu, Chun Yuan, Jian Guo, Sinan Du, Yiyan Qi, Zhengzhuo Xu","submitted_at":"2024-09-05T06:41:02Z","abstract_excerpt":"Automatic chart understanding is crucial for content comprehension and document parsing. Multimodal Large Language Models (MLLMs) have demonstrated remarkable capabilities in chart understanding through domain-specific alignment and fine-tuning. However, current MLLMs still struggle to provide faithful data and reliable analysis only based on charts. To address it, we propose ChartMoE, which employs the Mixture of Expert (MoE) architecture to replace the traditional linear projector to bridge the modality gap. Specifically, we train several linear connectors through distinct alignment tasks, w"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.03277","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/2409.03277/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":"2409.03277","created_at":"2026-07-05T10:30:56.325473+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.03277v3","created_at":"2026-07-05T10:30:56.325473+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.03277","created_at":"2026-07-05T10:30:56.325473+00:00"},{"alias_kind":"pith_short_12","alias_value":"QOBZJVEDG772","created_at":"2026-07-05T10:30:56.325473+00:00"},{"alias_kind":"pith_short_16","alias_value":"QOBZJVEDG772IENW","created_at":"2026-07-05T10:30:56.325473+00:00"},{"alias_kind":"pith_short_8","alias_value":"QOBZJVED","created_at":"2026-07-05T10:30:56.325473+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.03142","citing_title":"Disentangling Visual and Factual Correctness in LVLMs' Visualization Literacy","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29808","citing_title":"Making Multimodal LLMs Reliable Chart Data Extractors: A Benchmark and Training Framework","ref_index":77,"is_internal_anchor":false},{"citing_arxiv_id":"2512.19173","citing_title":"CycleChart: A Unified Consistency-Based Learning Framework for Bidirectional Chart Understanding and Generation","ref_index":43,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QOBZJVEDG772IENWJ23EMS4FUP","json":"https://pith.science/pith/QOBZJVEDG772IENWJ23EMS4FUP.json","graph_json":"https://pith.science/api/pith-number/QOBZJVEDG772IENWJ23EMS4FUP/graph.json","events_json":"https://pith.science/api/pith-number/QOBZJVEDG772IENWJ23EMS4FUP/events.json","paper":"https://pith.science/paper/QOBZJVED"},"agent_actions":{"view_html":"https://pith.science/pith/QOBZJVEDG772IENWJ23EMS4FUP","download_json":"https://pith.science/pith/QOBZJVEDG772IENWJ23EMS4FUP.json","view_paper":"https://pith.science/paper/QOBZJVED","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.03277&json=true","fetch_graph":"https://pith.science/api/pith-number/QOBZJVEDG772IENWJ23EMS4FUP/graph.json","fetch_events":"https://pith.science/api/pith-number/QOBZJVEDG772IENWJ23EMS4FUP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QOBZJVEDG772IENWJ23EMS4FUP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QOBZJVEDG772IENWJ23EMS4FUP/action/storage_attestation","attest_author":"https://pith.science/pith/QOBZJVEDG772IENWJ23EMS4FUP/action/author_attestation","sign_citation":"https://pith.science/pith/QOBZJVEDG772IENWJ23EMS4FUP/action/citation_signature","submit_replication":"https://pith.science/pith/QOBZJVEDG772IENWJ23EMS4FUP/action/replication_record"}},"created_at":"2026-07-05T10:30:56.325473+00:00","updated_at":"2026-07-05T10:30:56.325473+00:00"}