AI research papers.

arXiv:2512.23304v2 Announce Type: replace-cross Abstract: Multimodal Large Language Models (LLMs) introduce an emerging paradigm for medical imaging by interpreting scans through the lens of extensive clinical knowledge,…

arXiv:2609.26015v2 Announce Type: replace Abstract: Instruction-guided video editing requires executing a requested change while preserving the source video's appearance and temporal structure.

arXiv:2610.39016v1 Announce Type: cross Abstract: Compositional generalization remains a persistent limitation of transformer language models: even when a model learns to combine known facts correctly in-distribution (ID…

Evaluating Microsoft Copilot at enterprise scale requires test data that captures the full complexity of real organizations, yet real enterprise data contains sensitive information that cannot be used for evaluation unde…

Short-video search needs compact query rewriters that recover users' personalized intent from ambiguous queries under strict latency constraints.

Host Sowbhagya Suresh shares some of the latest news and cutting-edge research from across Microsoft—from understanding the challenges facing coding agents to uncovering how people use AI at work and improving extended-r…

While frontier vision-language models demonstrate strong capabilities in user interface (UI) navigation, their high inference costs prohibit everyday use.

Human cognition flexibly draws on a shared perceptual foundation to understand environments, attend to referenced entities, reason about spatial relationships, and act.

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 distribution into the model, leaving it una…

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 layer for agents that read, locate, and cit…

Whether models trained on different modalities converge to the same representation of the world is central to AI.

Formula-Driven Supervised Learning (FDSL) enables vision pre-training without real images, but existing methods cast pre-training as classification on discretized labels, discarding continuity in the generative process.

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 reasoning, and more agents.

One-step and few-step text-to-image generation must accurately render the details specified in the prompt despite limited opportunities for refinement.

Video diffusion transformers generate high-quality videos, but the quadratic cost of 3D full attention dominates their inference.

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 directly.

Recent advances in video world models have enabled the generation of increasingly realistic and interactive virtual environments, opening new possibilities for player-customizable gameplay.

Weight-normalized (WN) training is emerging as a compelling alternative to conventional weight decay (WD), with recent studies reporting substantial pretraining loss gains.

Vision-Language-Action (VLA) models have demonstrated promising capabilities in robotic manipulation by integrating visual perception, language understanding, and action generation.

We present PhysInsert, a framework for physically plausible reference-based object insertion that jointly adapts objects and their surrounding scenes.

Retrieval-augmented generation commonly serves separate models for reranking and answering, duplicating model residency and passage processing.

Foundation models are typically trained in stages: pretraining, midtraining, supervised fine-tuning (SFT), and reinforcement learning (RL), with changes in both data and learning algorithms.

The hair of stylized characters in games and avatars is built from hair cards: artist-placed ribbons whose geometry is the visible hair, animated by spring bones.

Interactive world models generate visual content conditioned on user actions, yet maintaining spatial consistency over long-horizon exploration in real time remains challenging.

Pixel-space generative Transformers typically use one token per image patch, so capturing finer detail requires more tokens and makes global attention increasingly expensive.

Achieving the broad task generalization of embodied models in physical environments remains a central challenge in robot learning.

Quarterly earnings forecasting requires an agent to connect financial evidence to the income-statement line items that produce a company's reported earnings.

We present EasyProver, a Lean proving system built from recursive agents with isolated contexts, asynchronous search, parallel tactic probing, and an elastic elaboration service.

Multi-Token Prediction (MTP) is now a common ingredient in language-model pre-training, as it improves downstream performance and enables faster inference through speculative decoding.

Financial agents must form forecasts and investment judgments from decision-time evidence, with predictions evaluated against subsequent outcomes.

Producing long-form videos from narration requires interdependent decisions that remain coherent across shots.

Humans routinely consume multimodal content, where textual and visual information is naturally interleaved.

Recursive self-improvement requires models not only to acquire useful training experience, but also to decide when their current experience has ceased to be useful.

Recent advances in multimodal large language models have increased interest in interaction paradigms for video understanding.

Self-distillation offers a way to improve reasoning models when a suitable external teacher is unavailable.

arXiv:2604.25370v3 Announce Type: replace-cross Abstract: The release of GPT-image-2 by OpenAI marks a watershed moment in AI-generated imagery: the boundary between photographic reality and synthetic content has never b…

arXiv:2610.32326v1 Announce Type: new Abstract: Large Language Model (LLM) agents have enabled automated code evolution for algorithm discovery.

This paper introduces Quaternion Approximate Networks (QUAN), a novel deep learning framework that leverages quaternion algebra for rotation equivariant image classification and object detection.

Diffusion video models produce striking clips, but they cannot reliably render an exact formula, a correct chart, a rule-consistent game or narration synchronized to on-screen events.

Low-cost arms such as the SO-101 have become the most common entry point to robot learning, yet no released policy performs spatially constrained, language-conditioned tasks on them out of the box, and existing SO-101 be…

Video stylization typically transforms entire scenes, while local artistic editing requires precise region--style association, spatial control, and temporal consistency.

We propose Nemotron-Labs-Diffusion-Uni (NLD-Uni), a unified discrete diffusion large language model (DLLM) for visual understanding and generation tasks.

Unified conditional image generation requires a single model to preserve different aspects of a visual input while producing a target image that follows textual guidance.

Diffusion vision-language models (DVLMs) process their entire input at every denoising step.

Interactive world generation requires models to follow successive commands while preserving relevant history and action consequences.

Learning controllable world dynamics from heterogeneous interaction videos is challenging because robot embodiments use incompatible action spaces, while human videos lack native control labels.

Multimodal perception forms the foundation of multimodal intelligence.

Scaling supervised training for working agents requires diverse professional tasks that yield reliable expert demonstrations.

AI research agents can propose, execute, and improve programs that solve machine learning tasks.