{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:EZXEXS3AGFZFPOZHBFTXK4YZ6G","short_pith_number":"pith:EZXEXS3A","schema_version":"1.0","canonical_sha256":"266e4bcb60317257bb270967757319f1baf2058ec26228da1a00444062a68d5e","source":{"kind":"arxiv","id":"2403.12596","version":1},"attestation_state":"computed","paper":{"title":"Chart-based Reasoning: Transferring Capabilities from LLMs to VLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Abhanshu Sharma, Fangyu Liu, Gilles Baechler, Hassan Mansoor, Jindong Chen, Rahul Aralikatte, Victor Carbune","submitted_at":"2024-03-19T10:03:07Z","abstract_excerpt":"Vision-language models (VLMs) are achieving increasingly strong performance on multimodal tasks. However, reasoning capabilities remain limited particularly for smaller VLMs, while those of large-language models (LLMs) have seen numerous improvements. We propose a technique to transfer capabilities from LLMs to VLMs. On the recently introduced ChartQA, our method obtains state-of-the-art performance when applied on the PaLI3-5B VLM by \\citet{chen2023pali3}, while also enabling much better performance on PlotQA and FigureQA.\n  We first improve the chart representation by continuing the pre-trai"},"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":"2403.12596","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-03-19T10:03:07Z","cross_cats_sorted":[],"title_canon_sha256":"44f8185aed21d42f582685d6483ca74f6d002391dcd71e451f8b32d79be63369","abstract_canon_sha256":"7c6eeba4c40ced61e9a9f89d4794e19c040f7da5312a5df8315cc8cce4fb305f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:58:04.340820Z","signature_b64":"IRGIRinBy8t/kXe9Xa82fwbBrb8Luaqi54/v1FIEuCHBFEs+67aTQRIIxUcTAh7FQFqRbyOvnf0eUmYGkqIOBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"266e4bcb60317257bb270967757319f1baf2058ec26228da1a00444062a68d5e","last_reissued_at":"2026-07-05T07:58:04.340331Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:58:04.340331Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Chart-based Reasoning: Transferring Capabilities from LLMs to VLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Abhanshu Sharma, Fangyu Liu, Gilles Baechler, Hassan Mansoor, Jindong Chen, Rahul Aralikatte, Victor Carbune","submitted_at":"2024-03-19T10:03:07Z","abstract_excerpt":"Vision-language models (VLMs) are achieving increasingly strong performance on multimodal tasks. However, reasoning capabilities remain limited particularly for smaller VLMs, while those of large-language models (LLMs) have seen numerous improvements. We propose a technique to transfer capabilities from LLMs to VLMs. On the recently introduced ChartQA, our method obtains state-of-the-art performance when applied on the PaLI3-5B VLM by \\citet{chen2023pali3}, while also enabling much better performance on PlotQA and FigureQA.\n  We first improve the chart representation by continuing the pre-trai"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.12596","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/2403.12596/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":"2403.12596","created_at":"2026-07-05T07:58:04.340395+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.12596v1","created_at":"2026-07-05T07:58:04.340395+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.12596","created_at":"2026-07-05T07:58:04.340395+00:00"},{"alias_kind":"pith_short_12","alias_value":"EZXEXS3AGFZF","created_at":"2026-07-05T07:58:04.340395+00:00"},{"alias_kind":"pith_short_16","alias_value":"EZXEXS3AGFZFPOZH","created_at":"2026-07-05T07:58:04.340395+00:00"},{"alias_kind":"pith_short_8","alias_value":"EZXEXS3A","created_at":"2026-07-05T07:58:04.340395+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.11470","citing_title":"The Periodic Table of LLM Reasoning: A Structured Survey of Reasoning Paradigms, Methods, and Failure Modes","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2604.13041","citing_title":"TableNet A Large-Scale Table Dataset with LLM-Powered Autonomous","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2604.10973","citing_title":"CFMS: A Coarse-to-Fine Multimodal Synthesis Framework for Enhanced Tabular Reasoning","ref_index":11,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EZXEXS3AGFZFPOZHBFTXK4YZ6G","json":"https://pith.science/pith/EZXEXS3AGFZFPOZHBFTXK4YZ6G.json","graph_json":"https://pith.science/api/pith-number/EZXEXS3AGFZFPOZHBFTXK4YZ6G/graph.json","events_json":"https://pith.science/api/pith-number/EZXEXS3AGFZFPOZHBFTXK4YZ6G/events.json","paper":"https://pith.science/paper/EZXEXS3A"},"agent_actions":{"view_html":"https://pith.science/pith/EZXEXS3AGFZFPOZHBFTXK4YZ6G","download_json":"https://pith.science/pith/EZXEXS3AGFZFPOZHBFTXK4YZ6G.json","view_paper":"https://pith.science/paper/EZXEXS3A","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.12596&json=true","fetch_graph":"https://pith.science/api/pith-number/EZXEXS3AGFZFPOZHBFTXK4YZ6G/graph.json","fetch_events":"https://pith.science/api/pith-number/EZXEXS3AGFZFPOZHBFTXK4YZ6G/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EZXEXS3AGFZFPOZHBFTXK4YZ6G/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EZXEXS3AGFZFPOZHBFTXK4YZ6G/action/storage_attestation","attest_author":"https://pith.science/pith/EZXEXS3AGFZFPOZHBFTXK4YZ6G/action/author_attestation","sign_citation":"https://pith.science/pith/EZXEXS3AGFZFPOZHBFTXK4YZ6G/action/citation_signature","submit_replication":"https://pith.science/pith/EZXEXS3AGFZFPOZHBFTXK4YZ6G/action/replication_record"}},"created_at":"2026-07-05T07:58:04.340395+00:00","updated_at":"2026-07-05T07:58:04.340395+00:00"}