ST-Think: How Multimodal Large Language Models Reason About 4D Worlds from Ego-Centric Videos
This work addresses spatial-temporal reasoning for MLLMs in video analysis, providing a new benchmark and method, but it is incremental as it builds on existing MLLM and reasoning techniques.
The paper tackles the problem of whether multimodal large language models (MLLMs) can understand 4D worlds from egocentric videos by introducing Ego-ST Bench, a benchmark with over 5,000 question-answer pairs, and proposing the ST-R1 training paradigm, which enhances performance through reverse thinking and reinforcement learning.
Humans excel at spatial-temporal reasoning, effortlessly interpreting dynamic visual events from an egocentric viewpoint. However, whether multimodal large language models (MLLMs) can similarly understand the 4D world remains uncertain. This paper explores multimodal spatial-temporal reasoning from an egocentric perspective, aiming to equip MLLMs with human-like reasoning capabilities. To support this objective, we introduce \textbf{Ego-ST Bench}, a novel benchmark containing over 5,000 question-answer pairs across four categories, systematically evaluating spatial, temporal, and integrated spatial-temporal reasoning. Additionally, we propose \textbf{ST-R1} training paradigm, a video-based reasoning model that incorporates reverse thinking into its reinforcement learning process, significantly enhancing performance. We combine long-chain-of-thought (long-CoT) supervised fine-tuning with Group Relative Policy Optimization (GRPO) reinforcement learning, achieving notable improvements with limited high-quality data. Ego-ST Bench and ST-R1 provide valuable insights and resources for advancing video-based spatial-temporal reasoning research.