AI research papers.

Large language models have achieved remarkable advances in complex reasoning.

Decoder hidden states from different layers of large language models (LLMs) often encode complementary information.

Visual representations are largely developed for understanding, yet what makes them useful for generation remains unclear.

Computer-use agents can perform increasingly complex GUI tasks, yet unfamiliar software and workflows requiring human domain knowledge remain challenging.

Creativity benchmarks for large language models (LLMs) vary widely in task and metric design.

Continual learning requires models to accumulate past experience and transfer it to new tasks.

Continual post-training faces a dual challenge of acquisition and retention: models must learn new capabilities while preserving prior knowledge.

We study when communication emerges between autonomous agents and what role an innate action selection architecture plays.

Terminal agents are moving beyond local operations toward completing real-world professional tasks.

Power sampling improves large language model (LLM) reasoning at test time by sampling from a sharpened version of the base model's distribution.

Efficient visual tokenization requires balancing representation compactness with the preservation of fine-grained image content.

On-policy distillation (OPD) trains a target policy on its own rollouts using token-level supervision from a stronger teacher, but typically relies on the teacher being substantially more capable than the target.

Visual grounding combines image--language understanding with geometric prediction, yet autoregressive multimodal models often express a box as a coordinate sequence.

Drag-based image editing provides an intuitive interface for spatially controlled image manipulation by allowing users to specify handle and target points.

As the production of software passes into the hands of software itself, coding agents are increasingly commissioned to build agents fitted to their users' own workflows.

Coding agents are increasingly used to build programs from scratch as software engineers do rather than to resolve issues within a repository or to generate a repository from a given requirements document.

Large language model (LLM) agents are becoming direct users of the open web.

On-policy distillation (OPD) provides teacher supervision on student-generated trajectories, but a single trajectory realizes only one continuation at each decision, limiting supervision coverage under a finite sampling…

Multi-teacher on-policy distillation (MOPD) extends on-policy distillation (OPD) by routing domain-specific data to specialized teachers, integrating their capabilities through dense token-level supervision on student-ge…

Professional agents increasingly work through command-line interfaces and multiple tools over shared workspaces containing files, messages, and business applications.

Evaluating large language models (LLMs) on AMLworld's smallest test splits is costly because each of over a million targets requires a separate graph prompt.

Attention Residuals (AttnRes) enables content-dependent retrieval over network depth, which can be viewed as single-stream softmax attention along the depth axis.

Driving world action models (WAMs) learn from paired trajectories and future observations.

When solving complex spatial reasoning tasks, ordinary humans naturally revisit the scenes from task-relevant viewpoints and mentally reconstruct the spatial layouts from those viewpoints.

Recursive self-improvement (RSI) aims to transform interaction experience into persistent improvements, yet external feedback is costly and world-model predictions may become unreliable as policies evolve.

Diffusion/flow models have emerged as the de facto paradigms for image generation, giving rise to a diverse family of modern generators with increasingly heterogeneous architectures and representations.

In this paper, we study physically grounded articulated 3D asset generation---moving the goal from reproducing how objects look to capturing how they move and function.

The rapid advancement of Large Language Models (LLMs) has driven their deployment toward general-purpose harness environments, where LLM-based agents operate in sandbox workspaces and interact directly with file systems…

We introduce OpenMobile-2, a near-frontier mobile agent with fully open training environments and recipes.

Understanding and tracking a dynamic world are fundamental to visual intelligence.

Diffusion language models (DLMs) offer parallel sampling for faster generation than autoregressive models, an increasingly important advantage as inference becomes a dominant computational workload.

Terminal agents solve tasks in software engineering and scientific research by executing commands in a terminal.

Coding agents such as Claude Code also follow text files that developers write for them: a project file, skills the agent may choose to open, and rules that load automatically when the agent reads matching files.

LLM agents' capabilities depend not only on their underlying models but also on their execution harnesses.

Pixel-space world models have recently emerged as a promising approach for GUI agents, predicting future screens prior to execution and subsequently re-perceiving them for action selection.

Automatic speech recognition produces words but drops the non-verbal content of a recording: the breaths, coughs, sighs and laughter of the speaker, and the horns, music or barking around them.

Whole-slide images (WSIs) are gigapixel-scale digital scans of histopathology tissue and serve as the primary basis for cancer diagnosis.

Hybrid reasoning models such as the Qwen3 series switch between a THINK mode and a NO-THINK mode through a few chat-template tokens.

Video diffusion models increasingly serve as priors for 4D generation, but the generated videos often lack geometric consistency.

World-action models inherit broad visual and physical priors from video pre-training.

Augmenting worker capabilities in the physical world through guidance has been a decades-long goal for AI agents.

Publishers and funders increasingly ask authors to disclose how AI tools contributed to a piece of research.

While learning dexterous manipulation from a single human video offers a promising alternative to costly robot demonstrations, many recent methods predominantly imitate demonstrated motions.

Training on broad, unfiltered data enables high-quality image generation, but can also teach models concepts that should not be generated; restricting the training data ensures safety, but reduces quality because example…

On-policy distillation (OPD) has become increasingly popular for transferring teacher capabilities to student models.

Reconstructing complete, scene-aligned 3D objects from casual images requires integrating sparse, uncertain observations and inferring surfaces hidden by occlusions.

Long-horizon tasks require LLM agents to continually draw on information from earlier interactions.

In an image latent space, the embeddings of high-resolution, natural, and sharp images form a manifold.

Work agents operate over digital artifacts to execute professional knowledge-intensive work, requiring training environments that support long-horizon interaction and trustworthy verification.

Learning from experience is essential for LLM agents to adapt to unfamiliar and dynmaic environments.