Autoregressive (AR) image generators are powerful in principle but inflexible in practice: training with a fixed raster-scan order hardcodes a single factorization of the image dis...
Safety alignment for Large Language Models is heavily focused on English with limited multilingual representation, particularly damaging low-resource language performance. Related ...
Predicting future physical states stands as a central challenge in embodied intelligence. Existing video world models leave 3D dynamics implicit, action-based formulations remain t...
Human cognition flexibly draws on a shared perceptual foundation to understand environments, attend to referenced entities, reason about spatial relationships, and act. Endowing ae...
We present PhysInsert, a framework for physically plausible reference-based object insertion that jointly adapts objects and their surrounding scenes. Existing methods often rely o...
Proactive agents must decide whether, when, and how far to assist without an explicit request. Existing evaluations often rely on automated scores or binary verdicts, offering limi...
A capable voice assistant must do more than recognize words; it must understand requests, reason over context, sustain conversations, and act through tools. We present Apriel Voice...
Large language model (LLM)-based multi-agent systems (MAS) coordinate specialized agents to solve complex tasks, turning each user request into interdependent model and tool operat...
Whether models trained on different modalities converge to the same representation of the world is central to AI. The Platonic Representation Hypothesis (PRH) conjectures that they...
On-policy Self-distillation (OPSD) has proven effective for post-training large language models by constructing a teacher conditioned on privileged references. However, existing ap...
While frontier vision-language models demonstrate strong capabilities in user interface (UI) navigation, their high inference costs prohibit everyday use. Conversely, smaller model...
Vision-language models can get individual judgments right while struggling to recover and consistently use the exact properties of a 3D object. To characterize this gap, we introdu...
Retrieval-augmented generation commonly serves separate models for reranking and answering, duplicating model residency and passage processing. We introduce Glance, which equips th...
Formula-Driven Supervised Learning (FDSL) enables vision pre-training without real images, but existing methods cast pre-training as classification on discretized labels, discardin...
Progress toward artificial general intelligence (AGI) has largely relied on scaling a resource-intensive executive, System-2, through larger models, more data and compute, longer r...
Achieving the broad task generalization of embodied models in physical environments remains a central challenge in robot learning. We study this challenge in World Action Models (W...
Vision-Language-Action (VLA) models have demonstrated promising capabilities in robotic manipulation by integrating visual perception, language understanding, and action generation...
Agentic image editors decompose complex requests into steps, but when a step fails, they mostly retry with the same narrow set of tools and a judge that scores each output by eye d...
We present Youtu-Parsing-Omni, an omni-modal parser that converts documents, images, audio, and video into structured records grounded in space and time, providing a perception lay...
Foundation models are typically trained in stages: pretraining, midtraining, supervised fine-tuning (SFT), and reinforcement learning (RL), with changes in both data and learning a...
Trajectory-level GRPO is a common approach for training multi-step GUI agents, but repeatedly rolling out complete trajectories from the task root can be inefficient: task-level re...
Effective computer-use agents (CUAs) should seamlessly combine visual GUI interaction with code and terminal execution within a single policy. We present Mage-VL-Cowork, a 4B hybri...
To succeed in real-world long-horizon software development, coding agents must satisfy multi-dimensional task specifications, such as interface compatibility and error handling, an...
The pursuit of fair and equitable AI in human-centric tasks is critically hindered by the inherent biases and privacy risks of real-world datasets, which offer imbalanced represent...