{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:ZFSV7CZXAF2EWADGBRWUQHPLP3","short_pith_number":"pith:ZFSV7CZX","schema_version":"1.0","canonical_sha256":"c9655f8b3701744b00660c6d481deb7edeed88da4f8237db9209e0c7a456d7e8","source":{"kind":"arxiv","id":"2508.11975","version":1},"attestation_state":"computed","paper":{"title":"Chart-CoCa: Self-Improving Chart Understanding of Vision LMs via Code-Driven Synthesis and Candidate-Conditioned Answering","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Gongyao Jiang, Qiong Luo","submitted_at":"2025-08-16T08:26:55Z","abstract_excerpt":"Vision Language Models (VLMs) often struggle with chart understanding tasks, particularly in accurate chart description and complex reasoning. Synthetic data generation is a promising solution, while usually facing the challenge of noise labels. To address this challenge, we first introduce a chart synthesis pipeline that generates aligned chart-question-answer triplets through code generation and execution, ensuring the reliability of synthetic data without human intervention. Furthermore, inspired by test-time scaling that increases inference budget and thereby improves performance, we desig"},"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":"2508.11975","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-16T08:26:55Z","cross_cats_sorted":[],"title_canon_sha256":"c9e923ea18e5237bab210ca5a4d95e85c76429a2aec1051553bbc5847a1b7007","abstract_canon_sha256":"d46786a18b923ff80758ba64b13ed6adeb2a50cc6c7e559a1b15fdff69081313"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:54:58.073903Z","signature_b64":"00eb8kVrjFi1g5lWyJ+PXXucsjjydUYDNeJ0EzziMpvIVrnwW9xBZ8VK8x4PpPQ/JLDwEsxrEgM430js5BqDDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c9655f8b3701744b00660c6d481deb7edeed88da4f8237db9209e0c7a456d7e8","last_reissued_at":"2026-07-05T11:54:58.073425Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:54:58.073425Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Chart-CoCa: Self-Improving Chart Understanding of Vision LMs via Code-Driven Synthesis and Candidate-Conditioned Answering","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Gongyao Jiang, Qiong Luo","submitted_at":"2025-08-16T08:26:55Z","abstract_excerpt":"Vision Language Models (VLMs) often struggle with chart understanding tasks, particularly in accurate chart description and complex reasoning. Synthetic data generation is a promising solution, while usually facing the challenge of noise labels. To address this challenge, we first introduce a chart synthesis pipeline that generates aligned chart-question-answer triplets through code generation and execution, ensuring the reliability of synthetic data without human intervention. Furthermore, inspired by test-time scaling that increases inference budget and thereby improves performance, we desig"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.11975","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/2508.11975/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":"2508.11975","created_at":"2026-07-05T11:54:58.073500+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.11975v1","created_at":"2026-07-05T11:54:58.073500+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.11975","created_at":"2026-07-05T11:54:58.073500+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZFSV7CZXAF2E","created_at":"2026-07-05T11:54:58.073500+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZFSV7CZXAF2EWADG","created_at":"2026-07-05T11:54:58.073500+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZFSV7CZX","created_at":"2026-07-05T11:54:58.073500+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2601.13606","citing_title":"ChartVerse: Scaling Chart Reasoning via Reliable Programmatic Synthesis from Scratch","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZFSV7CZXAF2EWADGBRWUQHPLP3","json":"https://pith.science/pith/ZFSV7CZXAF2EWADGBRWUQHPLP3.json","graph_json":"https://pith.science/api/pith-number/ZFSV7CZXAF2EWADGBRWUQHPLP3/graph.json","events_json":"https://pith.science/api/pith-number/ZFSV7CZXAF2EWADGBRWUQHPLP3/events.json","paper":"https://pith.science/paper/ZFSV7CZX"},"agent_actions":{"view_html":"https://pith.science/pith/ZFSV7CZXAF2EWADGBRWUQHPLP3","download_json":"https://pith.science/pith/ZFSV7CZXAF2EWADGBRWUQHPLP3.json","view_paper":"https://pith.science/paper/ZFSV7CZX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.11975&json=true","fetch_graph":"https://pith.science/api/pith-number/ZFSV7CZXAF2EWADGBRWUQHPLP3/graph.json","fetch_events":"https://pith.science/api/pith-number/ZFSV7CZXAF2EWADGBRWUQHPLP3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZFSV7CZXAF2EWADGBRWUQHPLP3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZFSV7CZXAF2EWADGBRWUQHPLP3/action/storage_attestation","attest_author":"https://pith.science/pith/ZFSV7CZXAF2EWADGBRWUQHPLP3/action/author_attestation","sign_citation":"https://pith.science/pith/ZFSV7CZXAF2EWADGBRWUQHPLP3/action/citation_signature","submit_replication":"https://pith.science/pith/ZFSV7CZXAF2EWADGBRWUQHPLP3/action/replication_record"}},"created_at":"2026-07-05T11:54:58.073500+00:00","updated_at":"2026-07-05T11:54:58.073500+00:00"}