{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:KKZF4YNQ3KYICDAS6MEUSYNIDJ","short_pith_number":"pith:KKZF4YNQ","schema_version":"1.0","canonical_sha256":"52b25e61b0dab0810c12f3094961a81a6b0f1753f33066f3eca1165613656d8f","source":{"kind":"arxiv","id":"2504.12316","version":1},"attestation_state":"computed","paper":{"title":"Data Metabolism: An Efficient Data Design Schema For Vision Language Model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.CL","authors_text":"Chenxi Sun, Fanheng Kong, Fuzheng Zhang, Hongzhi Zhang, Jiakang Wang, Jingyuan Zhang, Qi Wang, Xingguang Ji, Yahui Liu, Zhou Haonan","submitted_at":"2025-04-10T07:20:54Z","abstract_excerpt":"Data curation plays a crucial role in training powerful Visual Language Models (VLMs). In this work, we introduce the concept of Data Metabolism and present our data-centric framework to build VLMs throughout the development lifecycle. Starting from a standard model architecture, we discuss and provide insights into two crucial development steps: data curation and iteration, forming a closed-loop system that continuously improves model performance. We show a detailed codebook on how to process existing massive datasets and build user-specific data flywheel. As a demonstration, we release a VLM"},"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":"2504.12316","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-10T07:20:54Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"69e000bf06851143928102be38dd45b179939d1e08fd5a8086e67d723ccae85c","abstract_canon_sha256":"689532b646399c6ad156f4a78f38838de6230f0d1ba40593baedfc11f7fe2d32"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:50:13.968766Z","signature_b64":"G04sP5TK/SY7aL4oAZjpIV2ZSJNN4KWwNvxtlvatbWUGr7BEo6+9NkH0I2JoG129pmI8DxKK0B9WJv4IALJxBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"52b25e61b0dab0810c12f3094961a81a6b0f1753f33066f3eca1165613656d8f","last_reissued_at":"2026-07-05T10:50:13.968221Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:50:13.968221Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Data Metabolism: An Efficient Data Design Schema For Vision Language Model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.CL","authors_text":"Chenxi Sun, Fanheng Kong, Fuzheng Zhang, Hongzhi Zhang, Jiakang Wang, Jingyuan Zhang, Qi Wang, Xingguang Ji, Yahui Liu, Zhou Haonan","submitted_at":"2025-04-10T07:20:54Z","abstract_excerpt":"Data curation plays a crucial role in training powerful Visual Language Models (VLMs). In this work, we introduce the concept of Data Metabolism and present our data-centric framework to build VLMs throughout the development lifecycle. Starting from a standard model architecture, we discuss and provide insights into two crucial development steps: data curation and iteration, forming a closed-loop system that continuously improves model performance. We show a detailed codebook on how to process existing massive datasets and build user-specific data flywheel. As a demonstration, we release a VLM"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.12316","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/2504.12316/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":"2504.12316","created_at":"2026-07-05T10:50:13.968280+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.12316v1","created_at":"2026-07-05T10:50:13.968280+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.12316","created_at":"2026-07-05T10:50:13.968280+00:00"},{"alias_kind":"pith_short_12","alias_value":"KKZF4YNQ3KYI","created_at":"2026-07-05T10:50:13.968280+00:00"},{"alias_kind":"pith_short_16","alias_value":"KKZF4YNQ3KYICDAS","created_at":"2026-07-05T10:50:13.968280+00:00"},{"alias_kind":"pith_short_8","alias_value":"KKZF4YNQ","created_at":"2026-07-05T10:50:13.968280+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.20124","citing_title":"TUNA: Comprehensive Fine-grained Temporal Understanding Evaluation on Dense Dynamic Videos","ref_index":59,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KKZF4YNQ3KYICDAS6MEUSYNIDJ","json":"https://pith.science/pith/KKZF4YNQ3KYICDAS6MEUSYNIDJ.json","graph_json":"https://pith.science/api/pith-number/KKZF4YNQ3KYICDAS6MEUSYNIDJ/graph.json","events_json":"https://pith.science/api/pith-number/KKZF4YNQ3KYICDAS6MEUSYNIDJ/events.json","paper":"https://pith.science/paper/KKZF4YNQ"},"agent_actions":{"view_html":"https://pith.science/pith/KKZF4YNQ3KYICDAS6MEUSYNIDJ","download_json":"https://pith.science/pith/KKZF4YNQ3KYICDAS6MEUSYNIDJ.json","view_paper":"https://pith.science/paper/KKZF4YNQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.12316&json=true","fetch_graph":"https://pith.science/api/pith-number/KKZF4YNQ3KYICDAS6MEUSYNIDJ/graph.json","fetch_events":"https://pith.science/api/pith-number/KKZF4YNQ3KYICDAS6MEUSYNIDJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KKZF4YNQ3KYICDAS6MEUSYNIDJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KKZF4YNQ3KYICDAS6MEUSYNIDJ/action/storage_attestation","attest_author":"https://pith.science/pith/KKZF4YNQ3KYICDAS6MEUSYNIDJ/action/author_attestation","sign_citation":"https://pith.science/pith/KKZF4YNQ3KYICDAS6MEUSYNIDJ/action/citation_signature","submit_replication":"https://pith.science/pith/KKZF4YNQ3KYICDAS6MEUSYNIDJ/action/replication_record"}},"created_at":"2026-07-05T10:50:13.968280+00:00","updated_at":"2026-07-05T10:50:13.968280+00:00"}