[2404.10141] ANCHOR: LLM-driven Subject Conditioning for Text-to-Image Synthesis Skip to main content Search Submit Donate Log in Search arXiv Press Enter to search · Advanced search Computer Science > Computer Vision and Pattern Recognition arXiv:2404.10141 (cs) [Submitted on 15 Apr 2024 ( v1 ), last revised 25 Apr 2026 (this version, v2)] Title: ANCHOR: LLM-driven Subject Conditioning for Text-to-Image Synthesis Authors: Aashish Anantha Ramakrishnan , Sharon X. Huang , Dongwon Lee View a PDF of the paper titled ANCHOR: LLM-driven Subject Conditioning for Text-to-Image Synthesis, by Aashish Anantha Ramakrishnan and 1 other authors View PDF HTML (experimental) Abstract: Text-to-image (T2I) models have achieved remarkable progress in high-quality image synthesis, yet most benchmarks rely on simple, self-contained prompts, failing to capture the complexity of real-world captions. Human-written captions often involve multiple interacting subjects, rich contextual references, and abstractive phrasing, conditions under which current image-text encoders like CLIP struggle. To systematically study these deficiencies, we introduce ANCHOR, a large-scale dataset of 70K+ abstractive captions sourced from five major news media organizations. Analysis with ANCHOR reveals persistent failures in multi-subject understanding, context reasoning, and nuanced grounding. Motivated by these challenges, we propose Subject-Aware Fine-tuning (SAFE), which uses Large Language Models (LLMs) to extract key subjects and enhance their representation at the embedding-level. Experiments with contemporary models show that SAFE significantly improves image-caption consistency and human preference alignment, serving as a practical and scalable solution. Comments: Accepted to The 64th Annual Meeting of the Association for Computational Linguistics (ACL) 2026 Subjects: Computer Vision and Pattern Recognition (cs.CV) ; Computation and Language (cs.CL); Multimedia (cs.MM) MSC classes: 65D19 ACM classes: I.2.7; I.4.0 Cite as: arXiv:2404.10141 [cs.CV] (or arXiv:2404.10141v2 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2404.10141 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Aashish Anantha Ramakrishnan [ view email ] [v1] Mon, 15 Apr 2024 21:19:10 UTC (19,759 KB) [v2] Sat, 25 Apr 2026 04:41:00 UTC (16,500 KB) Full-text links: Access Paper: View a PDF of the paper titled ANCHOR: LLM-driven Subject Conditioning for Text-to-Image Synthesis, by Aashish Anantha Ramakrishnan and 1 other authors View PDF HTML (experimental) TeX Source view license Current browse context: cs.CV < prev | next > new | recent | 2024-04 Change to browse by: cs cs.CL cs.MM References & Citations NASA ADS Google Scholar Semantic Scholar export BibTeX citation Loading... BibTeX formatted citation × loading... Data provided by: Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer ( What is the Explorer? ) Connected Papers Toggle Connected Papers ( What is Connected Papers? ) Litmaps Toggle Litmaps ( What is Litmaps? ) scite.ai Toggle scite Smart Citations ( What are Smart Citations? ) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv ( What is alphaXiv? ) Links to Code Toggle CatalyzeX Code Finder for Papers ( What is CatalyzeX? ) DagsHub Toggle DagsHub ( What is DagsHub? ) GotitPub Toggle Gotit.pub ( What is GotitPub? ) Huggingface Toggle Hugging Face ( What is Huggingface? ) ScienceCast Toggle ScienceCast ( What is ScienceCast? ) Demos Demos Replicate Toggle Replicate ( What is Replicate? ) Spaces Toggle Hugging Face Spaces ( What is Spaces? ) Spaces Toggle TXYZ.AI ( What is TXYZ.AI? ) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower ( What are Influence Flowers? ) Core recommender toggle CORE Recommender ( What is CORE? ) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs . Which authors of this paper are endorsers? | Disable MathJax ( What is MathJax? ) We gratefully acknowledge support from our major funders , member institutions , , and all contributors. About · Help · Contact · Subscribe · Copyright · Privacy · Accessibility · Operational Status (opens in new tab) Major funding support from