{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:A66RZ2TCA5ER6VFTIF72EWEDPR","short_pith_number":"pith:A66RZ2TC","schema_version":"1.0","canonical_sha256":"07bd1cea6207491f54b3417fa258837c74e6d384617fb9fecdb78d2cdf783189","source":{"kind":"arxiv","id":"2509.02466","version":1},"attestation_state":"computed","paper":{"title":"TeRA: Rethinking Text-guided Realistic 3D Avatar Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hao Zhu, Jiawei Zhang, Li Wang, Xinxin Zuo, Xun Cao, Yanwen Wang, Yifei Zeng, Yiyu Zhuang","submitted_at":"2025-09-02T16:20:20Z","abstract_excerpt":"In this paper, we rethink text-to-avatar generative models by proposing TeRA, a more efficient and effective framework than the previous SDS-based models and general large 3D generative models. Our approach employs a two-stage training strategy for learning a native 3D avatar generative model. Initially, we distill a decoder to derive a structured latent space from a large human reconstruction model. Subsequently, a text-controlled latent diffusion model is trained to generate photorealistic 3D human avatars within this latent space. TeRA enhances the model performance by eliminating slow iter"},"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":"2509.02466","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-09-02T16:20:20Z","cross_cats_sorted":[],"title_canon_sha256":"99b8a4a61d4aca2b99e2ebc7efacb0dbdb5ca1f0258e7d7781bd5f677e615614","abstract_canon_sha256":"d3f2ff90bd846e35a730bb744056f61eb9db3fca89949f3b2558548c2a48ec42"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:04:19.302027Z","signature_b64":"+jh1264DNatFVUNHYu+BS5MnmloABQCsFl//GeWh6uuzSXWFGVxfEUOZirNf6fT6s+XAYTaDUQA07qlREiXbCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"07bd1cea6207491f54b3417fa258837c74e6d384617fb9fecdb78d2cdf783189","last_reissued_at":"2026-07-05T12:04:19.301529Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:04:19.301529Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TeRA: Rethinking Text-guided Realistic 3D Avatar Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hao Zhu, Jiawei Zhang, Li Wang, Xinxin Zuo, Xun Cao, Yanwen Wang, Yifei Zeng, Yiyu Zhuang","submitted_at":"2025-09-02T16:20:20Z","abstract_excerpt":"In this paper, we rethink text-to-avatar generative models by proposing TeRA, a more efficient and effective framework than the previous SDS-based models and general large 3D generative models. Our approach employs a two-stage training strategy for learning a native 3D avatar generative model. Initially, we distill a decoder to derive a structured latent space from a large human reconstruction model. Subsequently, a text-controlled latent diffusion model is trained to generate photorealistic 3D human avatars within this latent space. TeRA enhances the model performance by eliminating slow iter"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.02466","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/2509.02466/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":"2509.02466","created_at":"2026-07-05T12:04:19.301590+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.02466v1","created_at":"2026-07-05T12:04:19.301590+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.02466","created_at":"2026-07-05T12:04:19.301590+00:00"},{"alias_kind":"pith_short_12","alias_value":"A66RZ2TCA5ER","created_at":"2026-07-05T12:04:19.301590+00:00"},{"alias_kind":"pith_short_16","alias_value":"A66RZ2TCA5ER6VFT","created_at":"2026-07-05T12:04:19.301590+00:00"},{"alias_kind":"pith_short_8","alias_value":"A66RZ2TC","created_at":"2026-07-05T12:04:19.301590+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.02883","citing_title":"Information-Regularized Constrained Inversion for Stable Avatar Editing from Sparse Supervision","ref_index":32,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/A66RZ2TCA5ER6VFTIF72EWEDPR","json":"https://pith.science/pith/A66RZ2TCA5ER6VFTIF72EWEDPR.json","graph_json":"https://pith.science/api/pith-number/A66RZ2TCA5ER6VFTIF72EWEDPR/graph.json","events_json":"https://pith.science/api/pith-number/A66RZ2TCA5ER6VFTIF72EWEDPR/events.json","paper":"https://pith.science/paper/A66RZ2TC"},"agent_actions":{"view_html":"https://pith.science/pith/A66RZ2TCA5ER6VFTIF72EWEDPR","download_json":"https://pith.science/pith/A66RZ2TCA5ER6VFTIF72EWEDPR.json","view_paper":"https://pith.science/paper/A66RZ2TC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.02466&json=true","fetch_graph":"https://pith.science/api/pith-number/A66RZ2TCA5ER6VFTIF72EWEDPR/graph.json","fetch_events":"https://pith.science/api/pith-number/A66RZ2TCA5ER6VFTIF72EWEDPR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/A66RZ2TCA5ER6VFTIF72EWEDPR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/A66RZ2TCA5ER6VFTIF72EWEDPR/action/storage_attestation","attest_author":"https://pith.science/pith/A66RZ2TCA5ER6VFTIF72EWEDPR/action/author_attestation","sign_citation":"https://pith.science/pith/A66RZ2TCA5ER6VFTIF72EWEDPR/action/citation_signature","submit_replication":"https://pith.science/pith/A66RZ2TCA5ER6VFTIF72EWEDPR/action/replication_record"}},"created_at":"2026-07-05T12:04:19.301590+00:00","updated_at":"2026-07-05T12:04:19.301590+00:00"}