{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:2YR67UG4OBYRU4OB4YWVDA5FMG","short_pith_number":"pith:2YR67UG4","schema_version":"1.0","canonical_sha256":"d623efd0dc70711a71c1e62d5183a561a63612a5c1d3eed0ccdd07d7f6516743","source":{"kind":"arxiv","id":"2309.15807","version":1},"attestation_state":"computed","paper":{"title":"Emu: Enhancing Image Generation Models Using Photogenic Needles in a Haystack","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Abhimanyu Dubey, Abhishek Kadian, Chih-Yao Ma, Devi Parikh, Dhruv Mahajan, Filip Radenovic, Jialiang Wang, Ji Hou, Kunpeng Li, Matthew Yu, Mitesh Kumar Singh, Peizhao Zhang, Peter Vajda, Roshan Sumbaly, Rui Wang, Sam Tsai, Simon Vandenhende, Simran Motwani, Vignesh Ramanathan, Vladan Petrovic, Xiaofang Wang, Xiaoliang Dai, Yi Wen, Yiwen Song, Yue Zhao, Zijian He","submitted_at":"2023-09-27T17:30:19Z","abstract_excerpt":"Training text-to-image models with web scale image-text pairs enables the generation of a wide range of visual concepts from text. However, these pre-trained models often face challenges when it comes to generating highly aesthetic images. This creates the need for aesthetic alignment post pre-training. In this paper, we propose quality-tuning to effectively guide a pre-trained model to exclusively generate highly visually appealing images, while maintaining generality across visual concepts. Our key insight is that supervised fine-tuning with a set of surprisingly small but extremely visually"},"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":"2309.15807","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-09-27T17:30:19Z","cross_cats_sorted":[],"title_canon_sha256":"1598ddf9070062ecb3698ed97fd73a08575a250c38336c2139c4a82c77f72a9c","abstract_canon_sha256":"2dd8f80eabbf1fafd9cce1aac1fa2095ee7096f0e197b96950c6be3a770f4d77"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:54:57.960960Z","signature_b64":"GgXxX5D9ordj/q+8knNzrFfN7GF+5cXbXeepOXonDDdJnwxECb8r5V/9uzefUVxekHqH/VKTpTNlww8fY1ChAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d623efd0dc70711a71c1e62d5183a561a63612a5c1d3eed0ccdd07d7f6516743","last_reissued_at":"2026-07-05T06:54:57.960349Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:54:57.960349Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Emu: Enhancing Image Generation Models Using Photogenic Needles in a Haystack","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Abhimanyu Dubey, Abhishek Kadian, Chih-Yao Ma, Devi Parikh, Dhruv Mahajan, Filip Radenovic, Jialiang Wang, Ji Hou, Kunpeng Li, Matthew Yu, Mitesh Kumar Singh, Peizhao Zhang, Peter Vajda, Roshan Sumbaly, Rui Wang, Sam Tsai, Simon Vandenhende, Simran Motwani, Vignesh Ramanathan, Vladan Petrovic, Xiaofang Wang, Xiaoliang Dai, Yi Wen, Yiwen Song, Yue Zhao, Zijian He","submitted_at":"2023-09-27T17:30:19Z","abstract_excerpt":"Training text-to-image models with web scale image-text pairs enables the generation of a wide range of visual concepts from text. However, these pre-trained models often face challenges when it comes to generating highly aesthetic images. This creates the need for aesthetic alignment post pre-training. In this paper, we propose quality-tuning to effectively guide a pre-trained model to exclusively generate highly visually appealing images, while maintaining generality across visual concepts. Our key insight is that supervised fine-tuning with a set of surprisingly small but extremely visually"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.15807","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/2309.15807/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":"2309.15807","created_at":"2026-07-05T06:54:57.960427+00:00"},{"alias_kind":"arxiv_version","alias_value":"2309.15807v1","created_at":"2026-07-05T06:54:57.960427+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.15807","created_at":"2026-07-05T06:54:57.960427+00:00"},{"alias_kind":"pith_short_12","alias_value":"2YR67UG4OBYR","created_at":"2026-07-05T06:54:57.960427+00:00"},{"alias_kind":"pith_short_16","alias_value":"2YR67UG4OBYRU4OB","created_at":"2026-07-05T06:54:57.960427+00:00"},{"alias_kind":"pith_short_8","alias_value":"2YR67UG4","created_at":"2026-07-05T06:54:57.960427+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":15,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2603.14936","citing_title":"Bridging the Intention-Expression Gap: Aligning Multi-Dimensional Preferences via Hierarchical Relevance Feedback in Text-to-Image Diffusion","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15684","citing_title":"ElasticDiT: Efficient Diffusion Transformers via Elastic Architecture and Sparse Attention for High-Resolution Image Generation on Mobile Devices","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2505.22394","citing_title":"PacTure: Efficient PBR Texture Generation on Packed Views with Visual Autoregressive Models","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2502.10248","citing_title":"Step-Video-T2V Technical Report: The Practice, Challenges, and Future of Video Foundation Model","ref_index":290,"is_internal_anchor":false},{"citing_arxiv_id":"2310.19512","citing_title":"VideoCrafter1: Open Diffusion Models for High-Quality Video Generation","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2408.11039","citing_title":"Transfusion: Predict the Next Token and Diffuse Images with One Multi-Modal Model","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2604.25636","citing_title":"Refinement via Regeneration: Enlarging Modification Space Boosts Image Refinement in Unified Multimodal Models","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2604.24952","citing_title":"Learning from Noisy Preferences: A Semi-Supervised Learning Approach to Direct Preference Optimization","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2403.05135","citing_title":"ELLA: Equip Diffusion Models with LLM for Enhanced Semantic Alignment","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04494","citing_title":"Towards General Preference Alignment: Diffusion Models at Nash Equilibrium","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2410.13720","citing_title":"Movie Gen: A Cast of Media Foundation Models","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2501.03575","citing_title":"Cosmos World Foundation Model Platform for Physical AI","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18562","citing_title":"AnchorSeg: Language Grounded Query Banks for Reasoning Segmentation","ref_index":194,"is_internal_anchor":false},{"citing_arxiv_id":"2604.16552","citing_title":"Co-generation of Layout and Shape from Text via Autoregressive 3D Diffusion","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04609","citing_title":"Advancing Aesthetic Image Generation via Composition Transfer","ref_index":3,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2YR67UG4OBYRU4OB4YWVDA5FMG","json":"https://pith.science/pith/2YR67UG4OBYRU4OB4YWVDA5FMG.json","graph_json":"https://pith.science/api/pith-number/2YR67UG4OBYRU4OB4YWVDA5FMG/graph.json","events_json":"https://pith.science/api/pith-number/2YR67UG4OBYRU4OB4YWVDA5FMG/events.json","paper":"https://pith.science/paper/2YR67UG4"},"agent_actions":{"view_html":"https://pith.science/pith/2YR67UG4OBYRU4OB4YWVDA5FMG","download_json":"https://pith.science/pith/2YR67UG4OBYRU4OB4YWVDA5FMG.json","view_paper":"https://pith.science/paper/2YR67UG4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2309.15807&json=true","fetch_graph":"https://pith.science/api/pith-number/2YR67UG4OBYRU4OB4YWVDA5FMG/graph.json","fetch_events":"https://pith.science/api/pith-number/2YR67UG4OBYRU4OB4YWVDA5FMG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2YR67UG4OBYRU4OB4YWVDA5FMG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2YR67UG4OBYRU4OB4YWVDA5FMG/action/storage_attestation","attest_author":"https://pith.science/pith/2YR67UG4OBYRU4OB4YWVDA5FMG/action/author_attestation","sign_citation":"https://pith.science/pith/2YR67UG4OBYRU4OB4YWVDA5FMG/action/citation_signature","submit_replication":"https://pith.science/pith/2YR67UG4OBYRU4OB4YWVDA5FMG/action/replication_record"}},"created_at":"2026-07-05T06:54:57.960427+00:00","updated_at":"2026-07-05T06:54:57.960427+00:00"}