{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:JATOCAGPVKD4PPXRDNDWP5ZP7J","short_pith_number":"pith:JATOCAGP","schema_version":"1.0","canonical_sha256":"4826e100cfaa87c7bef11b4767f72ffa6feb3061489992489fc816289bc7c913","source":{"kind":"arxiv","id":"2503.10582","version":2},"attestation_state":"computed","paper":{"title":"VisualWebInstruct: Scaling up Multimodal Instruction Data through Web Search","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.CV","authors_text":"Bo Li, Jiachen Li, Kai Zou, Ping Nie, Wenhu Chen, Xiang Yue, Yiming Jia","submitted_at":"2025-03-13T17:32:48Z","abstract_excerpt":"Vision-Language Models have made significant progress on many perception-focused tasks. However, their progress on reasoning-focused tasks remains limited due to the lack of high-quality and diverse training data. In this work, we aim to address the scarcity of reasoning-focused multimodal datasets. We propose VisualWebInstruct, a novel approach that leverages search engines to create a diverse and high-quality dataset spanning multiple disciplines, including mathematics, physics, finance, and chemistry, etc. Starting with a meticulously selected set of 30,000 seed images, we employ Google Ima"},"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":"2503.10582","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-03-13T17:32:48Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"4a2a28a75e5bfd90bbfea7eac9907cd3a739db6b67c86f40c4deeffce8210dfd","abstract_canon_sha256":"f799a3881c2bcfe44a94fbeb45824d7e426645c3f51095593451005e737c1aea"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:31:57.110883Z","signature_b64":"TYlm9nAFWRulHUpB7QvNdqWqJrYjfgfirQo4O/JgdhIzcFih/2fCu0hI9xkwtnLpexFllb0ZwovFfUtrtsjJCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4826e100cfaa87c7bef11b4767f72ffa6feb3061489992489fc816289bc7c913","last_reissued_at":"2026-07-05T10:31:57.110248Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:31:57.110248Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VisualWebInstruct: Scaling up Multimodal Instruction Data through Web Search","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.CV","authors_text":"Bo Li, Jiachen Li, Kai Zou, Ping Nie, Wenhu Chen, Xiang Yue, Yiming Jia","submitted_at":"2025-03-13T17:32:48Z","abstract_excerpt":"Vision-Language Models have made significant progress on many perception-focused tasks. However, their progress on reasoning-focused tasks remains limited due to the lack of high-quality and diverse training data. In this work, we aim to address the scarcity of reasoning-focused multimodal datasets. We propose VisualWebInstruct, a novel approach that leverages search engines to create a diverse and high-quality dataset spanning multiple disciplines, including mathematics, physics, finance, and chemistry, etc. Starting with a meticulously selected set of 30,000 seed images, we employ Google Ima"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.10582","kind":"arxiv","version":2},"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/2503.10582/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":"2503.10582","created_at":"2026-07-05T10:31:57.110311+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.10582v2","created_at":"2026-07-05T10:31:57.110311+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.10582","created_at":"2026-07-05T10:31:57.110311+00:00"},{"alias_kind":"pith_short_12","alias_value":"JATOCAGPVKD4","created_at":"2026-07-05T10:31:57.110311+00:00"},{"alias_kind":"pith_short_16","alias_value":"JATOCAGPVKD4PPXR","created_at":"2026-07-05T10:31:57.110311+00:00"},{"alias_kind":"pith_short_8","alias_value":"JATOCAGP","created_at":"2026-07-05T10:31:57.110311+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.23543","citing_title":"VeriEvol: Scaling Multimodal Mathematical Reasoning via Verifiable Evol-Instruct","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2606.18216","citing_title":"Zone of Proximal Policy Optimization: Teacher in Prompts, Not Gradients","ref_index":105,"is_internal_anchor":false},{"citing_arxiv_id":"2505.16933","citing_title":"LLaDA-V: Large Language Diffusion Models with Visual Instruction Tuning","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2509.18154","citing_title":"MiniCPM-V 4.5: Cooking Efficient MLLMs via Architecture, Data, and Training Recipe","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14054","citing_title":"Bad Seeing or Bad Thinking? Rewarding Perception for Multimodal Reasoning","ref_index":45,"is_internal_anchor":false},{"citing_arxiv_id":"2604.11627","citing_title":"POINTS-Long: Adaptive Dual-Mode Visual Reasoning in MLLMs","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06247","citing_title":"SALLIE: Safeguarding Against Latent Language & Image Exploits","ref_index":10,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JATOCAGPVKD4PPXRDNDWP5ZP7J","json":"https://pith.science/pith/JATOCAGPVKD4PPXRDNDWP5ZP7J.json","graph_json":"https://pith.science/api/pith-number/JATOCAGPVKD4PPXRDNDWP5ZP7J/graph.json","events_json":"https://pith.science/api/pith-number/JATOCAGPVKD4PPXRDNDWP5ZP7J/events.json","paper":"https://pith.science/paper/JATOCAGP"},"agent_actions":{"view_html":"https://pith.science/pith/JATOCAGPVKD4PPXRDNDWP5ZP7J","download_json":"https://pith.science/pith/JATOCAGPVKD4PPXRDNDWP5ZP7J.json","view_paper":"https://pith.science/paper/JATOCAGP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.10582&json=true","fetch_graph":"https://pith.science/api/pith-number/JATOCAGPVKD4PPXRDNDWP5ZP7J/graph.json","fetch_events":"https://pith.science/api/pith-number/JATOCAGPVKD4PPXRDNDWP5ZP7J/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JATOCAGPVKD4PPXRDNDWP5ZP7J/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JATOCAGPVKD4PPXRDNDWP5ZP7J/action/storage_attestation","attest_author":"https://pith.science/pith/JATOCAGPVKD4PPXRDNDWP5ZP7J/action/author_attestation","sign_citation":"https://pith.science/pith/JATOCAGPVKD4PPXRDNDWP5ZP7J/action/citation_signature","submit_replication":"https://pith.science/pith/JATOCAGPVKD4PPXRDNDWP5ZP7J/action/replication_record"}},"created_at":"2026-07-05T10:31:57.110311+00:00","updated_at":"2026-07-05T10:31:57.110311+00:00"}