3D Gaussian avatars support fast rendering, however, their real-time animation is often challenged by the costly neural…
LLMs are increasingly applied to cybersecurity workflows, where they are expected to translate analysts' intent into too…
Building reliable robot capabilities across diverse tasks requires substantial human effort to develop and maintain skil…
What makes great scientists great? Even as AI systems start to make progress on open problems, scientists remain far ahe…
High-resolution 3D generation increasingly relies on voxel latents and multi-stage pipelines that first predict active s…
We show that multimodal models possess strong reasoning abilities and that an appropriate harness can unlock their poten…
Several state-of-the-art methods for online reinforcement learning in continuous control improve policies using action g…
Discrete diffusion language models offer a compelling alternative to autoregressive generation for tasks demanding bidir…
Distribution Matching Distillation (DMD) trains a few-step student from the difference between separately estimated targ…
Molecular learning models are strongly shaped by their underlying representations. Yet standard sequential and graph for…
Quasi-Newton (QN) methods have long been among the most effective methods for large-scale unconstrained convex optimizat…
Current controllable video generation systems often rely on 2D motion trajectories or sparse drag signals for object mot…
Robots operating in the physical world will increasingly need to coordinate with other robots, particularly in manipulat…
Vision-language models (VLMs) and vision-language-action models (VLAs) have recently driven rapid progress in general-pu…
Large language model (LLM) agents increasingly rely on persistent external sources to solve sequences of knowledge-inten…
Introducing new capabilities to frontier models has long been the goal of posttraining, which predominantly employs supe…
Unsupervised anomaly detection (UAD) methods for brain MRI are ranked by a single score, yet that score rests on choices…
We explore catastrophic forgetting in the context of large pre-trained models. By considering forgetting as a geometric…
Real-world enterprise data science and analytics workflows require reasoning across dozens of tables, performing statist…
On-policy self-distillation has recently emerged as an effective approach for improving language-model reasoning by supe…