Search for: All records

We introduce RIPT-VLA, a simple and scalable reinforcement-learning-based interactive post-training paradigm that fine-tunes pretrained Vision-Language-Action (VLA) models using only sparse binary success rewards. Existing VLA training pipelines rely heavily on offline expert demonstration data and supervised imitation, limiting their ability to adapt to new tasks and environments under low-data regimes. RIPT-VLA addresses this by enabling interactive post-training with a stable policy optimization algorithm based on dynamic rollout sampling and leave-one-out advantage estimation. RIPT-VLA has the following characteristics. First, it applies to various VLA models, resulting in an improvement on the lightweight QueST model by 21.2%, and the 7B OpenVLA-OFT model to an unprecedented 97.5% success rate. Second, it is computationally efficient and data-efficient: with only one demonstration, RIPT-VLA enables an unworkable SFT model (4%) to succeed with a 97% success rate within 15 iterations. Furthermore, we demonstrate that the policy learned by RIPT-VLA generalizes across different tasks and scenarios and is robust to the initial state context. These results highlight RIPT-VLA as a practical and effective paradigm for post-training VLA models through minimal supervision. 

Prior research has found that teacher reflection is important for both teachers’ professional development and students’ learning, but frequent, effective reflection faces many obstacles. Could an analytics-based reflection tool help address these barriers? Reflecto is a novel teacher-facing tool designed to support teachers in reflecting (after-the-fact) on their classroom practices during sessions when students are engaged with intelligent tutoring systems. Whereas many teacher analytics tools support real-time decision-making in class, Reflecto aims to promote teacher agency through out-of-class teacher-initiated, data-driven exploration of possible trends in their classroom practice. Unlike many other analytics tools, it combines learning analytics and teaching analytics.