2 papers
cs.CV2026
Overlapping Visual Grouping Without Semantic Priors
Teemu Saukkio, Hashem Haghbayan, Juha Plosila
Most computer-vision systems organize visual input toward a predefined interpretation, such as semantic categories, prompted regions, learned object-like representations, or a sing…
cs.LG2026
Can Explicit Physical Feasibility Benefit VLA Learning? An Empirical Study
Yubai Wei, Chen Wu, Hashem Haghbayan
Vision-Language-Action (VLA) models map multimodal inputs directly to robot actions and are typically trained through large-scale imitation learning. While this paradigm has shown…