{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:LDYST4ZSU37XE7QYOMXU32LH2U","short_pith_number":"pith:LDYST4ZS","schema_version":"1.0","canonical_sha256":"58f129f332a6ff727e18732f4de967d51b9af2284ff86d3f30a5ced5e884e24b","source":{"kind":"arxiv","id":"2406.18521","version":1},"attestation_state":"computed","paper":{"title":"CharXiv: Charting Gaps in Realistic Chart Understanding in Multimodal LLMs","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.CL","authors_text":"Alexis Chevalier, Danqi Chen, Haotian Liu, Howard Chen, Kaiqu Liang, Luxi He, Mengzhou Xia, Richard Zhu, Sadhika Malladi, Sanjeev Arora, Xindi Wu, Yitao Liu, Zirui Wang","submitted_at":"2024-06-26T17:50:11Z","abstract_excerpt":"Chart understanding plays a pivotal role when applying Multimodal Large Language Models (MLLMs) to real-world tasks such as analyzing scientific papers or financial reports. However, existing datasets often focus on oversimplified and homogeneous charts with template-based questions, leading to an over-optimistic measure of progress. We demonstrate that although open-source models can appear to outperform strong proprietary models on these benchmarks, a simple stress test with slightly different charts or questions can deteriorate performance by up to 34.5%. In this work, we propose CharXiv, a"},"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":"2406.18521","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-26T17:50:11Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"531199deafb8a1268646384aee85acfaf187cb6d1e2becffe6e48f68c2d96309","abstract_canon_sha256":"2e16402b3d921cd1ffc8789a782e57533d2903751f488c5743cb305229a2cf42"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:37:10.697604Z","signature_b64":"ifCg2hU7u7Wgq3nFzzbw/e3+g7JXbWVPc7EDqowMws6H+rQVb+LLQb03eSff61CJdOEYH8HvkSyXnPimZAKsAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"58f129f332a6ff727e18732f4de967d51b9af2284ff86d3f30a5ced5e884e24b","last_reissued_at":"2026-07-05T08:37:10.697105Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:37:10.697105Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CharXiv: Charting Gaps in Realistic Chart Understanding in Multimodal LLMs","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.CL","authors_text":"Alexis Chevalier, Danqi Chen, Haotian Liu, Howard Chen, Kaiqu Liang, Luxi He, Mengzhou Xia, Richard Zhu, Sadhika Malladi, Sanjeev Arora, Xindi Wu, Yitao Liu, Zirui Wang","submitted_at":"2024-06-26T17:50:11Z","abstract_excerpt":"Chart understanding plays a pivotal role when applying Multimodal Large Language Models (MLLMs) to real-world tasks such as analyzing scientific papers or financial reports. However, existing datasets often focus on oversimplified and homogeneous charts with template-based questions, leading to an over-optimistic measure of progress. We demonstrate that although open-source models can appear to outperform strong proprietary models on these benchmarks, a simple stress test with slightly different charts or questions can deteriorate performance by up to 34.5%. In this work, we propose CharXiv, a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.18521","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/2406.18521/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":"2406.18521","created_at":"2026-07-05T08:37:10.697159+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.18521v1","created_at":"2026-07-05T08:37:10.697159+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.18521","created_at":"2026-07-05T08:37:10.697159+00:00"},{"alias_kind":"pith_short_12","alias_value":"LDYST4ZSU37X","created_at":"2026-07-05T08:37:10.697159+00:00"},{"alias_kind":"pith_short_16","alias_value":"LDYST4ZSU37XE7QY","created_at":"2026-07-05T08:37:10.697159+00:00"},{"alias_kind":"pith_short_8","alias_value":"LDYST4ZS","created_at":"2026-07-05T08:37:10.697159+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":19,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.18216","citing_title":"Zone of Proximal Policy Optimization: Teacher in Prompts, Not Gradients","ref_index":132,"is_internal_anchor":false},{"citing_arxiv_id":"2606.17574","citing_title":"DeepInsight: A Unified Evaluation Infrastructure Across the Physical AI Stack","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01599","citing_title":"TRON: Targeted Rule-Verifiable Online Environments for Visual Reasoning RL","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09883","citing_title":"The Cartesian Shortcut: Re-evaluate Vision Reasoning in Polar Coordinate Space","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2605.27195","citing_title":"EpiCurveBench: Evaluating VLMs on Epidemic Curve Digitization","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.28714","citing_title":"IPO-Mine: A Toolkit and Dataset for Section-Structured Analysis of Long, Multimodal IPO Documents","ref_index":65,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23897","citing_title":"ETCHR: Editing To Clarify and Harness Reasoning","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2502.13923","citing_title":"Qwen2.5-VL Technical Report","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15691","citing_title":"SEED: Targeted Data Selection by Weighted Independent Set","ref_index":60,"is_internal_anchor":false},{"citing_arxiv_id":"2601.13606","citing_title":"ChartVerse: Scaling Chart Reasoning via Reliable Programmatic Synthesis from Scratch","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2604.03157","citing_title":"Chart-RL: Policy Optimization Reinforcement Learning for Enhanced Visual Reasoning in Chart Question Answering with Vision Language Models","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04172","citing_title":"GENFIG1: Visual Summaries of Scholarly Work as a Challenge for Vision-Language Models","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09883","citing_title":"The Cartesian Shortcut: Re-evaluate Vision Reasoning in Polar Coordinate Space","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19945","citing_title":"Visual Reasoning through Tool-supervised Reinforcement Learning","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2602.02276","citing_title":"Kimi K2.5: Visual Agentic Intelligence","ref_index":70,"is_internal_anchor":false},{"citing_arxiv_id":"2504.10479","citing_title":"InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models","ref_index":129,"is_internal_anchor":false},{"citing_arxiv_id":"2412.05271","citing_title":"Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling","ref_index":258,"is_internal_anchor":false},{"citing_arxiv_id":"2508.18265","citing_title":"InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency","ref_index":149,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04304","citing_title":"Hierarchical Visual Agent: Managing Contexts in Joint Image-Text Space for Advanced Chart Reasoning","ref_index":53,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LDYST4ZSU37XE7QYOMXU32LH2U","json":"https://pith.science/pith/LDYST4ZSU37XE7QYOMXU32LH2U.json","graph_json":"https://pith.science/api/pith-number/LDYST4ZSU37XE7QYOMXU32LH2U/graph.json","events_json":"https://pith.science/api/pith-number/LDYST4ZSU37XE7QYOMXU32LH2U/events.json","paper":"https://pith.science/paper/LDYST4ZS"},"agent_actions":{"view_html":"https://pith.science/pith/LDYST4ZSU37XE7QYOMXU32LH2U","download_json":"https://pith.science/pith/LDYST4ZSU37XE7QYOMXU32LH2U.json","view_paper":"https://pith.science/paper/LDYST4ZS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.18521&json=true","fetch_graph":"https://pith.science/api/pith-number/LDYST4ZSU37XE7QYOMXU32LH2U/graph.json","fetch_events":"https://pith.science/api/pith-number/LDYST4ZSU37XE7QYOMXU32LH2U/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LDYST4ZSU37XE7QYOMXU32LH2U/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LDYST4ZSU37XE7QYOMXU32LH2U/action/storage_attestation","attest_author":"https://pith.science/pith/LDYST4ZSU37XE7QYOMXU32LH2U/action/author_attestation","sign_citation":"https://pith.science/pith/LDYST4ZSU37XE7QYOMXU32LH2U/action/citation_signature","submit_replication":"https://pith.science/pith/LDYST4ZSU37XE7QYOMXU32LH2U/action/replication_record"}},"created_at":"2026-07-05T08:37:10.697159+00:00","updated_at":"2026-07-05T08:37:10.697159+00:00"}