超越指称表达：场景理解视觉 grounding
Beyond Referring Expressions: Scenario Comprehension Visual Grounding
AI总结：
本文提出RSC基准，探索基于场景的视觉 grounding，通过角色、意图和关系上下文推断目标，而非显式命名。ScenGround方法结合监督预热与难度感知强化学习，提升模型在复杂场景下的表现。
摘要：
现有视觉 grounding 基准主要评估图像区域与字面指称表达的对齐，模型常通过匹配显眼的命名类别成功。本文探索基于场景的视觉 grounding，目标需从角色、意图和关系上下文推断而非显式命名。引入Referring Scenario Comprehension (RSC)基准，查询为长段落文本，描述对象角色、用户目标和上下文线索，包括有意提及干扰对象需深度理解解决。每个实例标注可解释难度标签，暴露不同失败模式并支持细粒度分析。RSC包含约31k训练示例，4k领域内测试示例及3k分布外分割，含未见物体类别。进一步提出ScenGround，结合监督预热与难度感知强化学习，实验表明场景查询暴露当前模型系统性失败，而课程训练提升挑战片段表现并转移至标准基准。
英文摘要：
Existing visual grounding benchmarks primarily evaluate alignment between image regions and literal referring expressions, where models can often succeed by matching a prominent named category. We explore a complementary and more challenging setting of scenario-based visual grounding, where the target must be inferred from roles, intentions, and relational context rather than explicit naming. We introduce Referring Scenario Comprehension (RSC), a benchmark designed for this setting. The queries in this benchmark are paragraph-length texts that describe object roles, user goals, and contextual cues, including deliberate references to distractor objects that often require deep understanding to resolve. Each instance is annotated with interpretable difficulty tags for uniqueness, clutter, size, overlap, and position which expose distinct failure modes and support fine-grained analysis. RSC contains approximately 31k training examples, 4k in-domain test examples, and a 3k out-of-distribution split with unseen object categories. We further propose ScenGround, a curriculum reasoning method serving as a reference point for this setting, combining supervised warm-starting with difficulty-aware reinforcement learning. Experiments show that scenario-based queries expose systematic failures in current models that standard benchmarks do not reveal, and that curriculum training improves performance on challenging slices and transfers to standard benchmarks.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
