{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:CFGHH2PDSRPCEVKIXFF7GHELVQ","short_pith_number":"pith:CFGHH2PD","schema_version":"1.0","canonical_sha256":"114c73e9e3945e225548b94bf31c8bac23f07ef85d22b3628273b2e43694b938","source":{"kind":"arxiv","id":"2407.17453","version":2},"attestation_state":"computed","paper":{"title":"VILA$^2$: VILA Augmented VILA","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hongxu Yin, Jang Hyun Cho, Jan Kautz, Ligeng Zhu, Marco Pavone, Pavlo Molchanov, Song Han, Yan Wang, Yao Lu, Yunhao Fang","submitted_at":"2024-07-24T17:37:05Z","abstract_excerpt":"While visual language model architectures and training infrastructures advance rapidly, data curation remains under-explored where quantity and quality become a bottleneck. Existing work either crawls extra Internet data with a loose guarantee of quality or distills from black-box proprietary models, e.g., GPT-4V / Gemini that are API frequency and performance bounded. This work enables a VLM to improve itself via data enhancement, exploiting its generative nature. We introduce a simple yet effective VLM augmentation scheme that includes a self-augment step and a specialist-augment step to ite"},"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":"2407.17453","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-07-24T17:37:05Z","cross_cats_sorted":[],"title_canon_sha256":"4c608852344156e31ae672c30983e8aa4310e0874aaf5912fa03bd99f0fff31c","abstract_canon_sha256":"cce8570cfb2a68b5474c82bb3f5d3b414ea6349a952cefdc7e6b99e86330349e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:29:32.027425Z","signature_b64":"d4JlqlW3vaSIjE8ku1qPC2QSANkOuYoR6uTLnyTNn7Dd7W1JNiO/K9Ds72BBoHLotsml+YiSH+gmmNKdplbfBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"114c73e9e3945e225548b94bf31c8bac23f07ef85d22b3628273b2e43694b938","last_reissued_at":"2026-07-05T09:29:32.026709Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:29:32.026709Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VILA$^2$: VILA Augmented VILA","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hongxu Yin, Jang Hyun Cho, Jan Kautz, Ligeng Zhu, Marco Pavone, Pavlo Molchanov, Song Han, Yan Wang, Yao Lu, Yunhao Fang","submitted_at":"2024-07-24T17:37:05Z","abstract_excerpt":"While visual language model architectures and training infrastructures advance rapidly, data curation remains under-explored where quantity and quality become a bottleneck. Existing work either crawls extra Internet data with a loose guarantee of quality or distills from black-box proprietary models, e.g., GPT-4V / Gemini that are API frequency and performance bounded. This work enables a VLM to improve itself via data enhancement, exploiting its generative nature. We introduce a simple yet effective VLM augmentation scheme that includes a self-augment step and a specialist-augment step to ite"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.17453","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/2407.17453/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":"2407.17453","created_at":"2026-07-05T09:29:32.026795+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.17453v2","created_at":"2026-07-05T09:29:32.026795+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.17453","created_at":"2026-07-05T09:29:32.026795+00:00"},{"alias_kind":"pith_short_12","alias_value":"CFGHH2PDSRPC","created_at":"2026-07-05T09:29:32.026795+00:00"},{"alias_kind":"pith_short_16","alias_value":"CFGHH2PDSRPCEVKI","created_at":"2026-07-05T09:29:32.026795+00:00"},{"alias_kind":"pith_short_8","alias_value":"CFGHH2PD","created_at":"2026-07-05T09:29:32.026795+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.05552","citing_title":"Balancing Image Compression and Generation with Bootstrapped Tokenization","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2412.04468","citing_title":"NVILA: Efficient Frontier Visual Language Models","ref_index":73,"is_internal_anchor":false},{"citing_arxiv_id":"2502.04326","citing_title":"WorldSense: Evaluating Real-world Omnimodal Understanding for Multimodal LLMs","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2408.10188","citing_title":"LongVILA: Scaling Long-Context Visual Language Models for Long Videos","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2409.17146","citing_title":"Molmo and PixMo: Open Weights and Open Data for State-of-the-Art Vision-Language Models","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19193","citing_title":"How Far Are Video Models from True Multimodal Reasoning?","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CFGHH2PDSRPCEVKIXFF7GHELVQ","json":"https://pith.science/pith/CFGHH2PDSRPCEVKIXFF7GHELVQ.json","graph_json":"https://pith.science/api/pith-number/CFGHH2PDSRPCEVKIXFF7GHELVQ/graph.json","events_json":"https://pith.science/api/pith-number/CFGHH2PDSRPCEVKIXFF7GHELVQ/events.json","paper":"https://pith.science/paper/CFGHH2PD"},"agent_actions":{"view_html":"https://pith.science/pith/CFGHH2PDSRPCEVKIXFF7GHELVQ","download_json":"https://pith.science/pith/CFGHH2PDSRPCEVKIXFF7GHELVQ.json","view_paper":"https://pith.science/paper/CFGHH2PD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.17453&json=true","fetch_graph":"https://pith.science/api/pith-number/CFGHH2PDSRPCEVKIXFF7GHELVQ/graph.json","fetch_events":"https://pith.science/api/pith-number/CFGHH2PDSRPCEVKIXFF7GHELVQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CFGHH2PDSRPCEVKIXFF7GHELVQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CFGHH2PDSRPCEVKIXFF7GHELVQ/action/storage_attestation","attest_author":"https://pith.science/pith/CFGHH2PDSRPCEVKIXFF7GHELVQ/action/author_attestation","sign_citation":"https://pith.science/pith/CFGHH2PDSRPCEVKIXFF7GHELVQ/action/citation_signature","submit_replication":"https://pith.science/pith/CFGHH2PDSRPCEVKIXFF7GHELVQ/action/replication_record"}},"created_at":"2026-07-05T09:29:32.026795+00:00","updated_at":"2026-07-05T09:29:32.026795+00:00"}