Estimating task progress requires reasoning over long-horizon dynamics rather than recognizing static visual content. While modern Vision-Language Models (VLMs) excel at describing what is visible, it remains unclear whether they can infer how far a task has progressed from partial observations. To this end, we introduce Progress-Bench, a benchmark for systematically evaluating progress reasoning in VLMs. Beyond benchmarking, we further explore a human-inspired two-stage progress reasoning paradigm through both training-free prompting and training-based approach based on curated dataset ProgressLM-45K. Experiments on 14 VLMs show that most models are not yet ready for task progress estimation, exhibiting sensitivity to demonstration modality and viewpoint changes, as well as poor handling of unanswerable cases. While training-free prompting that enforces structured progress reasoning yields limited and model-dependent gains, the training-based ProgressLM-3B achieves consistent improvements even at a small model scale, despite being trained on a task set fully disjoint from the evaluation tasks. Further analyses reveal characteristic error patterns and clarify when and why progress reasoning succeeds or fails.

翻译：估计任务进展需要对长时程动态进行推理，而非识别静态视觉内容。尽管现代视觉语言模型（VLMs）在描述可见内容方面表现出色，但它们能否从部分观察中推断任务已进展多远仍不明确。为此，我们提出了Progress-Bench，一个用于系统评估VLMs中进展推理能力的基准。除基准测试外，我们进一步探索了一种受人类启发的两阶段进展推理范式，包括基于无训练提示的方法和基于精选数据集ProgressLM-45K的训练方法。在14个VLM上的实验表明，大多数模型尚未准备好进行任务进展估计，对演示模态和视角变化表现出敏感性，且难以处理不可回答的情况。尽管强制执行结构化进展推理的无训练提示方法带来了有限且模型依赖的性能提升，但基于训练的ProgressLM-3B即使在小模型规模下也实现了持续改进，尽管其训练任务集与评估任务完全不相交。进一步分析揭示了典型的错误模式，并阐明了进展推理何时及为何成功或失败。