AI research papers.

Captured human-object interactions provide rich supervision for humanoid loco-manipulation, but they are sparse, heterogeneous, and not directly executable by robots.

In this paper, we study how training data creates associations between the tokens at the start of a base model's response and the reasoning behavior that follows.

Low-precision execution can substantially accelerate reinforcement learning (RL) for large language models, but discrepancies between learner and sampler execution can destabilize policy optimization.

Whole-body tracking has become the interface through which operators drive humanoid robots, yet the references it consumes are recorded on level ground and carrying nothing, so the tracker is aware of neither the forces…

Diffusion large language models dLLMs) have emerged as a promising alternative to autoregressive language models through bidirectional diffusion-based token generation.

Mixture-of-Experts (MoE) transformers scale capacity by activating only a few experts per token, but this sparsity creates a hidden reliability problem: when routing is imperfect, load-balanced models may send tokens to…

``What are the irreducible conditions that are sufficient to produce an outcome?'' is one of the most common questions that recur across computation and science.

Wrapping an image generation model in an agentic harness can effectively boost Text-to-Image task performance: the harness can leverage memory, skills, workflow orchestration, result verification, and iterative refinemen…

Enterprise AI agents that share a memory store face two unaddressed risks: sensitive data can leak through legitimately computed results the requester could not derive, and departments can silently compute a same-named k…

Long-horizon LLM agents are frequently implemented using sparse mixture-of-experts (MoE) models, yet the co-design of agentic behavior and MoE structures remains underexplored.

On-policy reinforcement learning has become a central paradigm for improving the reasoning abilities of large language models.

Individual animal re-identification from camera-trap imagery is an instance retrieval problem central to non-invasive wildlife monitoring: a query image must retrieve the correct individual from a reference set of known…

Developers can build LLM agents by adapting third-party models through benign post-training.

Jev-style decision models return categorical probability distributions over predefined options without generating free-form text, enabling software systems to act on their outputs directly.

Open-ended discovery requires learning from repeated attempts while continuing to explore directions whose value is not yet apparent.

Recursive language-model agents decompose tasks and delegate subtasks to child instances of the same policy, forming a tree of work.

Robot agents must connect their intended actions to observed outcomes while retaining the context needed to revise their choices over repeated attempts.

Scaling robot data and model capacity has improved Vision-Language-Action (VLA) policies, but further progress is constrained by the high cost of robotic data.

Large language models (LLMs) are increasingly used as interactive agents and simulators, yet it remains unclear whether they can recover latent sequential structure beyond surface action frequencies.

Generating corner-case scenarios with appropriate adversity in a simulation environment is critical for testing an autonomous vehicle (AV) software stack's safety performance before deployment.

Recent GPU kernel optimization agents retain what they learn in knowledge bases or as distilled skills.

Continual skill evolution enables LLM agents to accumulate and refine reusable procedural knowledge from interaction experience without updating model parameters.

Generating a high-quality gaming world from a natural-language game intent requires joint reasoning about scene structure, spatial layout, gameplay objectives, interactive entities, and executable gameplay logic.

Evaluation budgets in agentic retrieval-augmented generation span questions, search trajectories, and repeated answers.

Test-time training (TTT) lets a model store information in its weights during inference.

Search agents repeatedly make short decisions about relevance, evidence sufficiency, and search actions.

Large language models (LLMs) progressively revise their predictions across Transformer layers, yet we typically observe only the final output, discarding the trajectory through which it is formed.

As agents rapidly evolve, existing benchmarks can become saturated, limiting their ability to distinguish capabilities and reveal remaining failure modes.

Long-term memory enables LLM-based agents to retain and reuse information across tasks and sessions, supporting personalization and long-horizon interactions.

Most vision-language-action models represent future motion as fixed-rate action chunks, tying temporal resolution and prediction horizon to a fixed output budget.

Recent advances in 3D Gaussian Splatting (3DGS) have achieved remarkable performance in novel view synthesis, yet deploying both static and dynamic Gaussian representations on resource-constrained mobile devices remains…

A large language model (LLM) agent solves long-horizon tasks through many reasoning-action turns, with one verification signal at termination.

Partnered human-humanoid interaction couples locomotion with continuous physical contact.

Natively training joint video-audio generation models at higher resolutions empowers them to learn richer visual details and sharper motion dynamics.

We introduce Loop Flow Transformers (LiFT), a family of looped generative models that scales computation by repeatedly applying a shared Diffusion Transformer (DiT) core, with only light changes to the standard architect…

We present Kandinsky 6.0 Video, a family of foundation diffusion models for synchronized text-to-audio-video generation, comprising Kandinsky 6.0 Video Lite (3B parameters) and Kandinsky 6.0 Video Pro (29B parameters).

Tool-using AI agents are increasingly deployed across enterprise software systems, yet widely used benchmarks primarily evaluate nominal task completion, conflating baseline planning competence with operational fault rec…

Vision-language models (VLMs) have achieved strong performance in multimodal reasoning, yet they remain prone to generating plausible but incorrect answers.

Routine complete blood counts (CBCs) could yield new biomarkers, but the private records needed to evaluate candidates cannot be shared with frontier language model agents that excel at discovery.

Vision-language-action (VLA) models face a timing gap between low-rate inference and high-rate robot execution.

World-action models (WAMs) augment robot policies with action-conditioned environment dynamics, yet existing approaches largely rely on future observation reconstruction or generic latent prediction and lack structured,…

Diffusion language models (DLMs) enable fast generation by predicting multiple tokens in parallel, but their practical adoption remains limited by a persistent quality gap relative to comparably sized autoregressive (AR)…

Skeletal motion is stored as every joint's transform at every frame, yet most of it is implied by the body rather than by what the motion is about.

LoRA-based continual learning methods mitigate catastrophic forgetting through various mechanisms, yet nearly all complement these with small learning rates as a heuristic to restrict gradient scaling magnitude.

LLM-based agents are increasingly capable of generating complex 3D structures, with the potential to reshape how objects are designed and realized in the physical world.

Self-evolving reasoning models learn from their own generated questions, yet repeated self-training can lead to performance collapse.

Efficient long-video understanding with vision-language models (VLMs) is often framed as selecting informative frames or visual tokens at a fixed native resolution.

Machine unlearning seeks to remove the influence of designated training data from a trained model without retraining from scratch.

On-policy distillation (OPD) has emerged as a widely used paradigm for post-training large language models, reducing the train--test mismatch of conventional distillation by supervising the student on its own generated t…

Many visual explanation methods in computer vision highlight pixel importance but struggle to link these low-level cues to semantically meaningful concepts, limiting their interpretability and trustworthiness.