OVOD-Agent：面向主动视觉推理与自演化检测的马尔可夫-班迪特框架
OVOD-Agent: A Markov-Bandit Framework for Proactive Visual Reasoning and Self-Evolving Detection
将开放词表检测的文本匹配过程重构为基于弱马尔可夫决策过程的智能体探索，利用Bandit模块在有限监督下生成探索信号，并通过自监督奖励模型闭环优化检测策略，为OVOD提供可复用的主动推理框架。
原文依据 we propose OVOD-Agent, which transforms passive category matching into proactive visual reasoning and self-evolving detection.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论开放词表目标检测（OVOD）旨在利用语义信息使检测器能够泛化到不同类别。尽管现有方法在大规模视觉-语言数据集上进行了预训练，但其推理仍局限于固定类别名称，导致多模态训练与单模态推理之间存在鸿沟。先前工作表明，改进文本表示可以显著提升OVOD性能，这表明文本空间仍未得到充分探索。为此，我们提出OVOD-Agent，将被动类别匹配转变为主动视觉推理与自演化检测。受思维链（CoT）范式启发，OVOD-Agent将文本优化过程扩展为具有显式动作的可解释视觉思维链。OVOD的轻量级特性使得基于LLM的管理不适用；相反，我们将视觉上下文转换建模为八个状态空间上的弱马尔可夫决策过程（w-MDP），这自然地表示了智能体的状态、记忆和交互动态。一个Bandit模块在有限监督下生成探索信号，帮助智能体关注不确定区域并调整其检测策略。我们进一步将马尔可夫转移矩阵与Bandit轨迹相结合，用于自监督奖励模型（RM）优化，形成从Bandit探索到RM学习的闭环。在COCO和LVIS上的实验表明，OVOD-Agent改进了现有OVOD基线，并在新类别上优于先前方法，展现出强大的泛化性和可扩展性。
英文原文摘要 1593 字符
Open-Vocabulary Object Detection (OVOD) aims to enable detectors to generalize across categories by leveraging semantic information. Although existing methods are pretrained on large vision-language datasets, their inference is still limited to fixed category names, creating a gap between multimodal training and unimodal inference. Previous work has shown that improving textual representation can significantly enhance OVOD performance, indicating that the textual space is still underexplored. To this end, we propose OVOD-Agent, which transforms passive category matching into proactive visual reasoning and self-evolving detection. Inspired by the Chain-of-Thought (CoT) paradigm, OVOD-Agent extends the textual optimization process into an interpretable Visual Chain of Thought with explicit actions. OVOD's lightweight nature makes LLM-based management unsuitable; instead, we model visual context transitions as a Weakly Markovian Decision Process (w-MDP) over eight state spaces, which naturally represents the agent's state, memory, and interaction dynamics. A Bandit module generates exploration signals under limited supervision, helping the agent focus on uncertain regions and adapt its detection policy. We further integrate Markov transition matrices with Bandit trajectories for self-supervised Reward Model (RM) optimization, forming a closed loop from Bandit exploration to RM learning. Experiments on COCO and LVIS show that OVOD-Agent improves existing OVOD baselines and outperforms prior methods on novel categories, demonstrating strong generalization and scalability.