AI research papers.

Large language model (LLM) agents increasingly rely on persistent external sources to solve sequences of knowledge-intensive tasks.

Precise control over camera and object motion is essential for professional video production.

How can a world model continuously observe regions beyond the actor's current view?

Multi-teacher on-policy distillation (MOPD) aims to combine the strengths of RL-trained teachers in a single student, but how teacher signals affect parameter changes remains underexplored.

We present OmniSeek, an agentic framework that transforms an Omni Large Language Model (Omni-LLM) into an active, multi-turn reasoning agent with native tool use.

Looped Transformers achieve parameter efficiency by repeatedly executing a shared block across recurrent loops.

Distribution Matching Distillation (DMD) trains a few-step student from the difference between separately estimated target and student scores, so it must keep an auxiliary diffusion model fitted to the student's evolving…

While Large Language Models (LLMs) have demonstrated striking capabilities on frontier mathematical problems, it remains unclear whether they possess the structural mathematical understanding underlying their solutions.

Discrete diffusion language models offer a compelling alternative to autoregressive generation for tasks demanding bidirectional reasoning and global constraint satisfaction.

We study test-time evolution for humanoid loco-manipulation: solving tasks that a controller was never trained for by repurposing its existing skills, improving from its own attempts, and retaining what it learns, withou…

High-resolution 3D generation increasingly relies on voxel latents and multi-stage pipelines that first predict active structure and then synthesize local geometry.

What makes great scientists great?

Programmable world models separate executable dynamics from visual generation, offering a promising foundation for next-generation game engines.

LLMs are increasingly applied to cybersecurity workflows, where they are expected to translate analysts' intent into tool invocations.

3D editing methods are usually tested on a single edit, yet an asset is built through a long sequence of revisions, each of which must implement the requested change while leaving everything else unchanged.

Existing Simulink benchmarks mainly evaluate whether generated models compile, execute, or resemble a reference model.

Computer-use agents need to reliably ground action targets in complex desktop scenes, where multiple applications, overlapping windows, and visually similar controls compete for attention.

Interactive world models are increasingly capable of generating environments and acting within them, yet deliberately editing an existing executable world remains underexplored.

Long-horizon compositional manipulation has become increasingly important for real-world robot deployment, where a single task involves multiple coordinated subtasks.

On-policy distillation (OPD) trains a student on the responses it generates.

Existing 3D large language models (LLMs) compromise on two fronts: they compress shapes into latent codebook indices or coordinate text, which removes spatial structure from what the model observes, and they acquire the…

Multi-agent systems (MAS) built from large language models coordinate specialized agents to tackle complex tasks, but effective workflows are difficult to design in advance.

Advances in experimental instrumentation and automation generate increasingly rich datasets, but turning experimental observations into microscopic understanding remains a bottleneck in scientific discovery.

Coding agents are extending their reach into the physical world by writing and executing robot control programs.

Recent years have seen the employment of a plethora of machine learning (ML) models in high-stakes domains, but they remain largely opaque to the practitioners who act on their predictions.

Typed decision models answer schema-constrained questions about a text in one forward pass and return probabilities meant to be thresholded.

World action models (WAMs) have emerged as a promising paradigm for robotic control by jointly predicting future visual dynamics and actions from an initial observation and instruction.

Long-video generation and world models have shown strong potential for interactive entertainment and embodied simulation by predicting future observations conditioned on user actions and historical memory.

A typed decision model answers a fixed question about an input by returning a probability for each of several caller-defined options.

Recovering executable CAD programs from 3D meshes is challenging due to the compositional nature of CAD construction and the interaction between discrete modeling choices and continuous parameters.

When a camera revisits a previously observed region, a video world model should reproduce what was there before.

Extreme space-weather events can damage power systems on Earth and degrade GPS accuracy and satellite operations.

arXiv:2609.35799v1 Announce Type: new Abstract: In July 2026, OpenAI's agents coordinated over channels outside their intended environment to breach Hugging Face's secured infrastructure.

Behavioral cloning becomes challenging when the same observation admits several valid actions.

Furnished floor plans support real-estate visualization, interior design, and architectural workflows, yet automatic furnishing remains challenged by limited real-world data and the need to satisfy interacting geometric…

Zero-shot text-to-speech (TTS) can reproduce an unseen speaker from a short reference recording, but typically entangles speaker identity and accent within the same reference.

Latent world models rely on representation geometry for planning, yet regularizing the latent marginal alone does not determine the state-to-state relationships used for action selection.

Multi-Head Latent Attention (MLA) enables expressive multi-head attention with compact caches for its content and decoupled RoPE paths, yet cache memory still scales linearly with context length and batch size.

Video world models require persistent scene memory to maintain consistency during long-horizon video generation.

Text-to-image diffusion models are personalized to a subject by DreamBooth fine-tuning on a handful of its images.

Large Vision-Language Models (LVLMs) have made remarkable progress across visual perception tasks, yet spatial reasoning remains a persistent weakness, especially for questions that require reasoning over visual space.

Large Vision-Language Models (LVLMs) commonly perform spatial reasoning through chain-of-thought (CoT), encoding intermediate reasoning as autoregressive sequences of discrete language tokens.

Modern multimodal models bring generation and understanding into a single unified system, which enables them to provide and learn from their own feedback.

Recent joint video-audio generation models have achieved strong semantic correspondence and temporal synchronization.

Patch retrieval, the task of finding the commit that fixes a known vulnerability, is the foundation of vulnerability management workflows, yet 60% to 63% of CVEs in the major advisory databases lack a patch link.

Autoregressive models are trained to predict a system's behavior one step at a time, and recursive generation allows the learned dynamics to unfold over long horizons.

Anthropic's Agent Skills package reusable procedural know-how for an LLM agent into SKILL.md directories, and open-source aggregations have grown past 230,000 skills, making selection rather than authoring the bottleneck…

Direct-decision models turn text into low-latency structured labels and scores, making them attractive for classification and automatic evaluation.

Long-horizon video generation requires models to effectively leverage an increasingly long generation history.