Vision-Language-Action Models (VLA) for robotics are trained by combining large language models with vision encoders and then fine-tuning them on various robot datasets; this allows generalization to new instructions, unseen objects, and distribution shifts. However, various real-world robot datasets mostly require human control, which makes scaling difficult. On the other hand, Internet video data offers… →

A primary feature of sophisticated language models is In-Context Learning (ICL), which allows the model to produce answers based on input instances without being specifically instructed on how to complete the task. In ICL, a few examples that show the intended behavior or pattern are shown to the model, which then applies this knowledge to… →

The discovery of new materials is crucial to addressing pressing global challenges such as climate change and advancements in next-generation computing. However, existing computational and experimental approaches face significant limitations in efficiently exploring the vast chemical space. While AI has emerged as a powerful tool for materials discovery, the lack of publicly available data and… →

The growing reliance on large language models for coding support poses a significant problem: how best to assess real-world impact on programmer productivity? Current approaches, such as static bench-marking based on datasets such as HumanEval, measure the correctness of the code but cannot capture the dynamic, human-in-the-loop interaction of real programming activity. With LLMs increasingly… →

In the rapidly evolving world of AI, challenges related to scalability, performance, and accessibility remain central to the efforts of research communities and open-source advocates. Issues such as the computational demands of large-scale models, the lack of diverse model sizes for different use cases, and the need to balance accuracy with efficiency are critical obstacles.… →

One of the most critical challenges of LLMs is how to align these models with human values and preferences, especially in generated texts. Most generated text outputs by models are inaccurate, biased, or potentially harmful—for example, hallucinations. This misalignment limits the potential usage of LLMs in real-world applications across domains such as education, health, and… →

Human beings possess innate extraordinary perceptual judgments, and when computer vision models are aligned with them, model’s performance can be improved manifold. Various attributes such as scene layout, subject location, camera pose, color, perspective, and semantics help us have a clear picture of the world and objects within. The alignment of vision models with visual… →

Visual and action data are interconnected in robotic tasks, forming a perception-action loop. Robots rely on control parameters for movement, while VFMs excel in processing visual data. However, a modality gap exists between visual and action data arising from the fundamental differences in their sensory modalities, abstraction levels, temporal dynamics, contextual dependence, and susceptibility to… →