{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:4MDLVEJ2GFD4BUWASSMU24P7RI","short_pith_number":"pith:4MDLVEJ2","schema_version":"1.0","canonical_sha256":"e306ba913a3147c0d2c094994d71ff8a0cfc7efc89dc8191fbb51a541705e7f5","source":{"kind":"arxiv","id":"2606.02580","version":1},"attestation_state":"computed","paper":{"title":"Thinking in Blender: Staged Executable Inverse Graphics with Vision-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Guangzhao He, Hadar Averbuch-Elor, Rundong Luo, Wei-Chiu Ma","submitted_at":"2026-06-01T17:59:53Z","abstract_excerpt":"Inverse graphics is a longstanding and highly underconstrained problem that seeks to reconstruct images as editable 3D scenes which can be rendered, relit, and manipulated. In this work, we investigate whether pretrained vision-language models (VLMs) can perform executable inverse graphics directly from a single image by reconstructing a scene as an editable Blender program, without relying on specialized 2D or 3D foundation models, differentiable rendering, or multi-view supervision. We introduce Staged Executable Inverse Graphics (SEIG), an agentic framework that reconstructs a 3D scene from"},"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":"2606.02580","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-06-01T17:59:53Z","cross_cats_sorted":[],"title_canon_sha256":"0c676e2247bece1c5f592eea042bc6335a810a319caae1e8d0fc171ab7ce968c","abstract_canon_sha256":"5aefc9205b50c2269756e78d07ef418802179ccba5eeab378841bc4886850a17"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T03:05:10.727292Z","signature_b64":"0HvzfeHV0Us+ckclqesqMlsMLlUMRLHae0bORAvwLYoqe2zpKhCgsihCgfqgBemPWM6TYnfVTmPErXKMEAHLDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e306ba913a3147c0d2c094994d71ff8a0cfc7efc89dc8191fbb51a541705e7f5","last_reissued_at":"2026-06-02T03:05:10.726872Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T03:05:10.726872Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Thinking in Blender: Staged Executable Inverse Graphics with Vision-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Guangzhao He, Hadar Averbuch-Elor, Rundong Luo, Wei-Chiu Ma","submitted_at":"2026-06-01T17:59:53Z","abstract_excerpt":"Inverse graphics is a longstanding and highly underconstrained problem that seeks to reconstruct images as editable 3D scenes which can be rendered, relit, and manipulated. In this work, we investigate whether pretrained vision-language models (VLMs) can perform executable inverse graphics directly from a single image by reconstructing a scene as an editable Blender program, without relying on specialized 2D or 3D foundation models, differentiable rendering, or multi-view supervision. We introduce Staged Executable Inverse Graphics (SEIG), an agentic framework that reconstructs a 3D scene from"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.02580","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/2606.02580/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":"2606.02580","created_at":"2026-06-02T03:05:10.726936+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.02580v1","created_at":"2026-06-02T03:05:10.726936+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.02580","created_at":"2026-06-02T03:05:10.726936+00:00"},{"alias_kind":"pith_short_12","alias_value":"4MDLVEJ2GFD4","created_at":"2026-06-02T03:05:10.726936+00:00"},{"alias_kind":"pith_short_16","alias_value":"4MDLVEJ2GFD4BUWA","created_at":"2026-06-02T03:05:10.726936+00:00"},{"alias_kind":"pith_short_8","alias_value":"4MDLVEJ2","created_at":"2026-06-02T03:05:10.726936+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/4MDLVEJ2GFD4BUWASSMU24P7RI","json":"https://pith.science/pith/4MDLVEJ2GFD4BUWASSMU24P7RI.json","graph_json":"https://pith.science/api/pith-number/4MDLVEJ2GFD4BUWASSMU24P7RI/graph.json","events_json":"https://pith.science/api/pith-number/4MDLVEJ2GFD4BUWASSMU24P7RI/events.json","paper":"https://pith.science/paper/4MDLVEJ2"},"agent_actions":{"view_html":"https://pith.science/pith/4MDLVEJ2GFD4BUWASSMU24P7RI","download_json":"https://pith.science/pith/4MDLVEJ2GFD4BUWASSMU24P7RI.json","view_paper":"https://pith.science/paper/4MDLVEJ2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.02580&json=true","fetch_graph":"https://pith.science/api/pith-number/4MDLVEJ2GFD4BUWASSMU24P7RI/graph.json","fetch_events":"https://pith.science/api/pith-number/4MDLVEJ2GFD4BUWASSMU24P7RI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4MDLVEJ2GFD4BUWASSMU24P7RI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4MDLVEJ2GFD4BUWASSMU24P7RI/action/storage_attestation","attest_author":"https://pith.science/pith/4MDLVEJ2GFD4BUWASSMU24P7RI/action/author_attestation","sign_citation":"https://pith.science/pith/4MDLVEJ2GFD4BUWASSMU24P7RI/action/citation_signature","submit_replication":"https://pith.science/pith/4MDLVEJ2GFD4BUWASSMU24P7RI/action/replication_record"}},"created_at":"2026-06-02T03:05:10.726936+00:00","updated_at":"2026-06-02T03:05:10.726936+00:00"}