PROGRESSLM: Towards Progress Reasoning in Vision-Language Models

TL;DR

This paper introduces Progress-Bench and ProgressLM-45K to evaluate and improve vision-language models' ability to reason about task progress, revealing current limitations and potential improvements through training-based methods.

Contribution

It presents a new benchmark and dataset for progress reasoning in VLMs, and demonstrates that training-based approaches can enhance progress estimation performance.

Findings

Most VLMs struggle with progress estimation and are sensitive to modality and viewpoint.

Training-free prompting yields limited gains and is model-dependent.

Training-based ProgressLM-3B improves progress reasoning even with small models.

Abstract

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.