{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:TOEKK7EECC3ORFFGLELK7QCYMT","short_pith_number":"pith:TOEKK7EE","schema_version":"1.0","canonical_sha256":"9b88a57c8410b6e894a65916afc05864eef0bd8a78716b9b546ecf186b55eb4f","source":{"kind":"arxiv","id":"2606.20364","version":1},"attestation_state":"computed","paper":{"title":"Judging to Improve: A De-biased VLM-as-3D-Judge Protocol for Single-Image 3D Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Ali Asaria, Deep Gandhi, Tony Salomone","submitted_at":"2026-06-18T15:25:57Z","abstract_excerpt":"A companion study established a de-biased, cross-model VLM-as-3D-judge that reliably ranks single-image-to-3D mesh quality where cheap geometry and CLIP proxies fall short. This paper asks: can that judge's preferences specialize a strong open generator, TRELLIS, on one asset class (furniture), cheaply and without human labels? Taking the judge from ranking to optimization is where the work lives. Pushing a VLM judge into the training and evaluation loop exposes failure modes ranking never triggered, so our contribution is an optimization-grade hardening of the judge: a training judge (Qwen2.5"},"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":"2606.20364","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-18T15:25:57Z","cross_cats_sorted":[],"title_canon_sha256":"90babc059ba88efaf2a568b5ba26b01f30aaa110d85d5cbf686a9e52af95bd4e","abstract_canon_sha256":"42a3f73d2dafd8d00b43b7b6ad5e9192d01e7953ab2f853c80b9cdf4a19c49da"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:13:09.948226Z","signature_b64":"a3zR3rPPOpOx1kbx1JBKZrIhAXUdsVsKK3DUj02SpvuIVJ6bog+pw0KmxsfBn6sMJW8q3qiy235x1EY7aXM8AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9b88a57c8410b6e894a65916afc05864eef0bd8a78716b9b546ecf186b55eb4f","last_reissued_at":"2026-06-19T16:13:09.947855Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:13:09.947855Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Judging to Improve: A De-biased VLM-as-3D-Judge Protocol for Single-Image 3D Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Ali Asaria, Deep Gandhi, Tony Salomone","submitted_at":"2026-06-18T15:25:57Z","abstract_excerpt":"A companion study established a de-biased, cross-model VLM-as-3D-judge that reliably ranks single-image-to-3D mesh quality where cheap geometry and CLIP proxies fall short. This paper asks: can that judge's preferences specialize a strong open generator, TRELLIS, on one asset class (furniture), cheaply and without human labels? Taking the judge from ranking to optimization is where the work lives. Pushing a VLM judge into the training and evaluation loop exposes failure modes ranking never triggered, so our contribution is an optimization-grade hardening of the judge: a training judge (Qwen2.5"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.20364","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/2606.20364/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":"2606.20364","created_at":"2026-06-19T16:13:09.947916+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.20364v1","created_at":"2026-06-19T16:13:09.947916+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.20364","created_at":"2026-06-19T16:13:09.947916+00:00"},{"alias_kind":"pith_short_12","alias_value":"TOEKK7EECC3O","created_at":"2026-06-19T16:13:09.947916+00:00"},{"alias_kind":"pith_short_16","alias_value":"TOEKK7EECC3ORFFG","created_at":"2026-06-19T16:13:09.947916+00:00"},{"alias_kind":"pith_short_8","alias_value":"TOEKK7EE","created_at":"2026-06-19T16:13:09.947916+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/TOEKK7EECC3ORFFGLELK7QCYMT","json":"https://pith.science/pith/TOEKK7EECC3ORFFGLELK7QCYMT.json","graph_json":"https://pith.science/api/pith-number/TOEKK7EECC3ORFFGLELK7QCYMT/graph.json","events_json":"https://pith.science/api/pith-number/TOEKK7EECC3ORFFGLELK7QCYMT/events.json","paper":"https://pith.science/paper/TOEKK7EE"},"agent_actions":{"view_html":"https://pith.science/pith/TOEKK7EECC3ORFFGLELK7QCYMT","download_json":"https://pith.science/pith/TOEKK7EECC3ORFFGLELK7QCYMT.json","view_paper":"https://pith.science/paper/TOEKK7EE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.20364&json=true","fetch_graph":"https://pith.science/api/pith-number/TOEKK7EECC3ORFFGLELK7QCYMT/graph.json","fetch_events":"https://pith.science/api/pith-number/TOEKK7EECC3ORFFGLELK7QCYMT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TOEKK7EECC3ORFFGLELK7QCYMT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TOEKK7EECC3ORFFGLELK7QCYMT/action/storage_attestation","attest_author":"https://pith.science/pith/TOEKK7EECC3ORFFGLELK7QCYMT/action/author_attestation","sign_citation":"https://pith.science/pith/TOEKK7EECC3ORFFGLELK7QCYMT/action/citation_signature","submit_replication":"https://pith.science/pith/TOEKK7EECC3ORFFGLELK7QCYMT/action/replication_record"}},"created_at":"2026-06-19T16:13:09.947916+00:00","updated_at":"2026-06-19T16:13:09.947916+00:00"}