{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:P26KD62LMFG3FYFU2MGEKBQZVZ","short_pith_number":"pith:P26KD62L","schema_version":"1.0","canonical_sha256":"7ebca1fb4b614db2e0b4d30c450619ae605528d9be8ed72ff525095c3fb1a402","source":{"kind":"arxiv","id":"2507.08513","version":2},"attestation_state":"computed","paper":{"title":"Advancing Multimodal LLMs by Large-Scale 3D Visual Instruction Dataset Generation","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.GR","authors_text":"Albert Y. C. Chen, Cheng-Hao Kuo, Daniel Aliaga, Liu He, Lu Xia, Min Sun, Sankalp Dayal, Shashwat Verma, Xiao Zeng, Yizhi Song","submitted_at":"2025-07-11T12:00:10Z","abstract_excerpt":"Multimodal Large Language Models (MLLMs) struggle with accurately capturing camera-object relations, especially for object orientation, camera viewpoint, and camera shots. This stems from the fact that existing MLLMs are trained on images with limited diverse camera-object relations and corresponding textual descriptions. To address this, we propose a synthetic generation pipeline to create large-scale 3D visual instruction datasets. Our framework takes 3D assets as input and uses rendering and diffusion-based image generation models to create photorealistic images preserving precise camera-ob"},"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":"2507.08513","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.GR","submitted_at":"2025-07-11T12:00:10Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"06a4561c42b678b0f10a42a00c9fc9391982b49b108c588e871b79e9b7e52b99","abstract_canon_sha256":"1b4241e172502782c0933b3213c0827c01df7196e560cb1144dce30251f4b1de"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:42:32.335879Z","signature_b64":"ZvRWMbxjxq5V3QvSY+vNDXqyCZUJWx51Y0axy/Z6zWbcS8/yvX1yvlPxQ1wC09eQvl7hY/OgoT/KEHtDgxJwAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7ebca1fb4b614db2e0b4d30c450619ae605528d9be8ed72ff525095c3fb1a402","last_reissued_at":"2026-07-05T11:42:32.335387Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:42:32.335387Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Advancing Multimodal LLMs by Large-Scale 3D Visual Instruction Dataset Generation","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.GR","authors_text":"Albert Y. C. Chen, Cheng-Hao Kuo, Daniel Aliaga, Liu He, Lu Xia, Min Sun, Sankalp Dayal, Shashwat Verma, Xiao Zeng, Yizhi Song","submitted_at":"2025-07-11T12:00:10Z","abstract_excerpt":"Multimodal Large Language Models (MLLMs) struggle with accurately capturing camera-object relations, especially for object orientation, camera viewpoint, and camera shots. This stems from the fact that existing MLLMs are trained on images with limited diverse camera-object relations and corresponding textual descriptions. To address this, we propose a synthetic generation pipeline to create large-scale 3D visual instruction datasets. Our framework takes 3D assets as input and uses rendering and diffusion-based image generation models to create photorealistic images preserving precise camera-ob"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.08513","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/2507.08513/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":"2507.08513","created_at":"2026-07-05T11:42:32.335462+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.08513v2","created_at":"2026-07-05T11:42:32.335462+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.08513","created_at":"2026-07-05T11:42:32.335462+00:00"},{"alias_kind":"pith_short_12","alias_value":"P26KD62LMFG3","created_at":"2026-07-05T11:42:32.335462+00:00"},{"alias_kind":"pith_short_16","alias_value":"P26KD62LMFG3FYFU","created_at":"2026-07-05T11:42:32.335462+00:00"},{"alias_kind":"pith_short_8","alias_value":"P26KD62L","created_at":"2026-07-05T11:42:32.335462+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/P26KD62LMFG3FYFU2MGEKBQZVZ","json":"https://pith.science/pith/P26KD62LMFG3FYFU2MGEKBQZVZ.json","graph_json":"https://pith.science/api/pith-number/P26KD62LMFG3FYFU2MGEKBQZVZ/graph.json","events_json":"https://pith.science/api/pith-number/P26KD62LMFG3FYFU2MGEKBQZVZ/events.json","paper":"https://pith.science/paper/P26KD62L"},"agent_actions":{"view_html":"https://pith.science/pith/P26KD62LMFG3FYFU2MGEKBQZVZ","download_json":"https://pith.science/pith/P26KD62LMFG3FYFU2MGEKBQZVZ.json","view_paper":"https://pith.science/paper/P26KD62L","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.08513&json=true","fetch_graph":"https://pith.science/api/pith-number/P26KD62LMFG3FYFU2MGEKBQZVZ/graph.json","fetch_events":"https://pith.science/api/pith-number/P26KD62LMFG3FYFU2MGEKBQZVZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/P26KD62LMFG3FYFU2MGEKBQZVZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/P26KD62LMFG3FYFU2MGEKBQZVZ/action/storage_attestation","attest_author":"https://pith.science/pith/P26KD62LMFG3FYFU2MGEKBQZVZ/action/author_attestation","sign_citation":"https://pith.science/pith/P26KD62LMFG3FYFU2MGEKBQZVZ/action/citation_signature","submit_replication":"https://pith.science/pith/P26KD62LMFG3FYFU2MGEKBQZVZ/action/replication_record"}},"created_at":"2026-07-05T11:42:32.335462+00:00","updated_at":"2026-07-05T11:42:32.335462+00:00"}