eprintid: 10192852 rev_number: 15 eprint_status: archive userid: 699 dir: disk0/10/19/28/52 datestamp: 2024-05-30 12:23:31 lastmod: 2024-11-04 14:45:24 status_changed: 2024-05-30 12:23:31 type: proceedings_section metadata_visibility: show sword_depositor: 699 creators_name: Gong, Jingzhi creators_name: Li, Sisi creators_name: D’Aloisio, Giordano creators_name: Ding, Zishuo creators_name: Yulong, Ye creators_name: Langdon, William B creators_name: Sarro, Federica title: GreenStableYolo: Optimizing Inference Time and Image Quality of Text-to-Image Generation ispublished: pub divisions: UCL divisions: B04 divisions: C05 divisions: F48 keywords: SBSE; ANN; GenAI; Text2Image; Stable Diffusion; Yolo note: This version is the author accepted manuscript. For information on re-use, please refer to the publisher's terms and conditions. abstract: Tuning the parameters and prompts for improving AI-based text-to-image generation has remained a substantial yet unaddressed challenge. Hence we introduce GreenStableYolo, which improves the parameters and prompts for Stable Diffusion to both reduce GPU inference time and increase image generation quality using NSGA-II and Yolo. Our experiments show that despite a relatively slight trade-off (18%) in image quality compared to StableYolo (which only considers image quality), GreenStableYolo achieves a substantial reduction in inference time (266% less) and a 526% higher hypervolume, thereby advancing the state-of-the-art for text-to-image generation. date: 2024-07-02 date_type: published publisher: Springer official_url: https://doi.org/10.1007/978-3-031-64573-0_7 full_text_type: other language: eng verified: verified_manual elements_id: 2279057 doi: 10.1007/978-3-031-64573-0_7 isbn_13: 978-3-031-64572-3 lyricists_name: Sarro, Federica lyricists_id: FSSAR72 actors_name: Sarro, Federica actors_id: FSSAR72 actors_role: owner full_text_status: restricted pres_type: paper series: Lecture Notes in Computer Science publication: International Symposium on Search Based Software Engineering volume: 14767 place_of_pub: Cham, Switzerland pagerange: 70-76 event_title: 16th International Symposium, SSBSE 2024 issn: 0302-9743 book_title: Search-Based Software Engineering: SSBSE 2024 editors_name: Jahangirova, Gunel editors_name: Khomh, Foutse citation: Gong, Jingzhi; Li, Sisi; D’Aloisio, Giordano; Ding, Zishuo; Yulong, Ye; Langdon, William B; Sarro, Federica; (2024) GreenStableYolo: Optimizing Inference Time and Image Quality of Text-to-Image Generation. In: Jahangirova, Gunel and Khomh, Foutse, (eds.) Search-Based Software Engineering: SSBSE 2024. (pp. pp. 70-76). Springer: Cham, Switzerland. document_url: https://discovery.ucl.ac.uk/id/eprint/10192852/1/GreenStableYolo.pdf