paper-with-me

홈 › Papers

RePlan: Reasoning-guided Region Planning for Complex Instruction-based Image Editing

2025-12-18 · Tianyuan Qu, Lei Ke, Xiaohang Zhan, Longxiang Tang, Yuqi Liu, Bohao Peng, Bei Yu, Dong Yu, Jiaya Jia arxiv

Instruction-based image editing enables natural-language control over visual modifications, yet existing models falter under Instruction-Visual Complexity (IV-Complexity), where intricate instructions meet cluttered or ambiguous scenes. We introduce RePlan (Region-aligned Planning), a plan-then-execute framework that couples a vision-language planner with a diffusion editor. The planner decomposes instructions via step-by-step reasoning and explicitly grounds them to target regions; the editor then applies changes using a training-free attention-region injection mechanism, enabling precise, parallel multi-region edits without iterative inpainting. To strengthen planning, we apply GRPO-based reinforcement learning using 1K instruction-only examples, yielding substantial gains in reasoning fidelity and format reliability. We further present IV-Edit, a benchmark focused on fine-grained grounding and knowledge-intensive edits. Across IV-Complex settings, RePlan consistently outperforms strong baselines trained on far larger datasets, improving regional precision and overall consistency. Our project page: https://replan-iv-edit.github.io

📄 PDF Abstract BibTeX arXiv:2512.16864

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement LearningImage Editing

Similar Papers 제목 키워드 기반

RePlan-Bot: Multi-Level Replanning for Embodied Instruction Following

2026-05-25 · Xicheng Gong, Guozheng Sun, Peiran Xu, Yadong Mu arxiv

Embodied instruction following (EIF) requires agents to understand and execute complex natural language commands within interactive 3D environments. Despite recent advances, existing methods often fail in long-horizon pl…

Instruction FollowingObject Localization

Scene Graph-Guided Proactive Replanning for Failure-Resilient Embodied Agent

2025-08-15 · Che Rin Yu, Daewon Chae, Dabin Seo, Sangwon Lee 외 arxiv

When humans perform everyday tasks, we naturally adjust our actions based on the current state of the environment. For instance, if we intend to put something into a drawer but notice it is closed, we open it first. Howe…

Plan Along the Way: Event-Triggered Foundation-Model Planning for TAMP Execution in Partially Observable Manipulation

2026-08-28 · Puru Ojha, Narendhiran Vijayakumar, Nav Singhal, Girish Varma 외 arxiv

Manipulation in partially observable environments requires planning under incomplete scene information. In such settings, an initially valid plan may execute successfully yet remain insufficient for task completion. Exis…

Motion Planning

Compositional Monte Carlo Tree Diffusion for Extendable Planning

2025-10-24 · Jaesik Yoon, Hyeonseo Cho, Sungjin Ahn arxiv

Monte Carlo Tree Diffusion (MCTD) integrates diffusion models with structured tree search to enable effective trajectory exploration through stepwise reasoning. However, MCTD remains fundamentally limited by training tra…

When Tools Fail: Benchmarking Dynamic Replanning and Anomaly Recovery in LLM Agents

2026-06-04 · Dongsheng Zhu, Xuchen Ma, Yucheng Shen, Xiang Li 외 arxiv

Existing benchmarks evaluate Tool-Integrated Reasoning (TIR) in LLMs on idealized ''happy paths'', largely overlooking real-world tool failures. We introduce ToolMaze, a benchmark for dynamic path discovery and error rec…