CountLoop: Training-Free High-Instance Image Generation via Iterative Agent Guidance

[2508.16644] CountLoop: Training-Free High-Instance Image Generation via Iterative Agent Guidance 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:2508.16644 (cs) [Submitted on 18 Aug 2025 ( v1 ), last revised 11 Apr 2026 (this version, v4)] Title: CountLoop: Training-Free High-Instance Image Generation via Iterative Agent Guidance Authors: Anindya Mondal , Ayan Banerjee , Sauradip Nag , Josep Llados , Xiatian Zhu , Anjan Dutta View a PDF of the paper titled CountLoop: Training-Free High-Instance Image Generation via Iterative Agent Guidance, by Anindya Mondal and 5 other authors View PDF HTML (experimental) Abstract: Diffusion models excel at photorealistic synthesis but struggle with precise object counts, especially in high-density settings. We introduce COUNTLOOP, a training-free framework that achieves precise instance control through iterative, structured feedback. Our method alternates between synthesis and evaluation: a VLM-based planner generates structured scene layouts, while a VLM-based critic provides explicit feedback on object counts, spatial arrangements, and visual quality to refine the layout iteratively. Instance-driven attention masking and cumulative attention composition further prevent semantic leakage, ensuring clear object separation even in densely occluded scenes. Evaluations on COCO-Count, T2I-CompBench, and two newly introduced high instance benchmarks show that COUNTLOOP reduces counting error by up to 57% and achieves the highest or comparable spatial quality scores across all benchmarks, while maintaining photorealism. Subjects: Computer Vision and Pattern Recognition (cs.CV) Cite as: arXiv:2508.16644 [cs.CV] (or arXiv:2508.16644v4 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2508.16644 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Ayan Banerjee [ view email ] [v1] Mon, 18 Aug 2025 11:28:02 UTC (453,961 KB) [v2] Sun, 7 Dec 2025 21:18:36 UTC (27,974 KB) [v3] Mon, 16 Mar 2026 23:45:34 UTC (35,394 KB) [v4] Sat, 11 Apr 2026 08:03:25 UTC (35,394 KB) Full-text links: Access Paper: View a PDF of the paper titled CountLoop: Training-Free High-Instance Image Generation via Iterative Agent Guidance, by Anindya Mondal and 5 other authors View PDF HTML (experimental) TeX Source view license Current browse context: cs.CV < prev | next > new | recent | 2025-08 Change to browse by: cs 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

Related documents

• ViFA-Council: Multi-Agent LLM Deliberation for Vietnamese Folk Art Generation#1071147
• Towards Controllable Image Generation through Representation-Conditioned Diffusion Models#229481
• Explorative Modeling: Unlocking a Third Pretraining Axis and End-to-End Generation#421570
• Versatile On-device Adaptation at the Edge by Unifying Few-shot, Zero-shot, Continual, and In-context Learning#422269
• Ecdysis: Efficient and Effective Training of Runtime Harnesses for LLM Agents#999371