AI research papers.

Progress toward general intelligence in the physical world requires scaling robot learning across diverse environments and interactions.

Video scene graph generation (VidSGG) aims to capture objects and their evolving relationships in videos.

Distilling pretrained video diffusion models into few-step autoregressive (AR) generators typically combines student initialization with distribution matching distillation (DMD).

KANLab is a browser-based visualization tool designed to interactively teach the core concepts of Kolmogorov-Arnold Networks (KAN), forked from the widely successful TensorFlow Playground.

Current video object removal methods suffer from prohibitive latency (minutes per video) and non-causal processing that prevents real-time interaction.

Effective social interaction requires more than fluent responses: agents must reach plans that fit the participants' needs and circumstances.

Autoregressive generation underlies real-time, causal video synthesis and is the backbone of video world models.

Long-horizon agents must coordinate reasoning, tool use, and environment feedback over extended interactions.

Feed-forward Gaussian reconstruction builds 3D scenes directly from input images for novel view rendering, but learning Gaussian attributes from predicted geometry for high-quality rendering under large viewpoint changes…

Agents driven by large language models are increasingly deployed on long-horizon tasks, where they use tools and modify real workspaces.

Context can improve large language model performance, but these gains may be lost after context removal.

Existing long-horizon agent benchmarks often emphasize final task success while providing limited assessment of intermediate execution validity.

Recursive self-improvement (RSI) in scientific research aims to improve agents' research capabilities through their own experimental experience.

Foundation models are driving a paradigm shift in robotic intelligence, moving beyond low-level action prediction to goal-directed orchestration of data collection, model training, task planning, and real-world execution…

The scaling relationship of compute and performance for language models (LMs) has been extensively studied; more recently, models are being designed to scale memory use, as well.

Chemical chain-of-thought can contain hallucinated structural and property claims.

Character Error Rate (CER), the standard metric for OCR and vision-language models (VLMs), charges a typo as much as a wrong name.

Speculative decoding accelerates autoregressive generation by drafting multiple tokens per step, yet feature-based drafters rely on target hidden features whose contextual emphasis may not align with the needs of subsequ…

Generative super-resolution (SR) can recover sharp detail in computed tomography (CT) images from severely downsampled observations, but the missing high-frequency content is not uniquely determined by the low-resolution…

On-policy self-distillation provides dense supervision for reasoning models via a privileged self-teacher.

Existing robot policy memories typically evolve with observation sequences, implicitly treating each update as the same amount of physical time.

Pretrained unimodal models offer a flexible route to multimodal prediction through collaboration, but this collaboration can become less accurate than a surviving unimodal model when modalities are missing.

Interactive video generation commonly relies on small bidirectional chunks to reduce response latency.

Long-form agentic video generation must turn a screenplay into shots that remain spatially and temporally consistent across nonconsecutive requests.

On-policy distillation (OPD) trains a student model on its own rollouts with token-level supervision from a teacher, assuming that the teacher can reach a correct answer from any prefix the student writes.

Long contexts enlarge key--value (KV) caches, increasing memory use and attention costs during generation.

Computer-use agents have demonstrated strong capabilities in interpreting screens and executing keyboard and mouse actions.

Scaling up transformers has produced models with broad capabilities, deployed across heterogeneous hardware and fluctuating workloads.

We present CommerceGen-Studio, an end-to-end framework for generating commercial-grade main images for e-commerce Stock Keeping Units (SKUs).

World-action models aim to turn visual anticipation into driving decisions, yet typically train action experts on trajectory representations defined independently of visual dynamics.

Vision-Language-Action (VLA) models are gaining attention in autonomous driving, inheriting commonsense priors from large language models (LLMs) and widening the navigation interface from a fixed command set to free-form…

Sparse autoencoders (SAEs) are typically trained on hundreds of millions to billions of activation tokens, yet little work has tested whether this scale is necessary.

Video generators serve as interactive worlds for exploration and simulation, yet realistic and visually appealing frames do not necessarily depict a consistent 3D world.

Humans interpret multimodal observations of the world and forecast how it will evolve.

Unified multimodal generation covers discrete, continuous, and hybrid paradigms.

Reasoning distillation requires chain-of-thought (CoT) data with correct answers, but obtaining such data efficiently remains challenging.

Despite their extensive pretraining knowledge and strong reasoning capabilities, large language models (LLMs) struggle to adaptively gather information in experimental design tasks.

Online reinforcement learning (RL) with verifiable rewards has driven rapid progress in mathematics, coding, and tool use, but scaling this success to GUI agents remains challenging due to the high cost of interactive en…

Decision models answer typed questions about an input with one probability per option, computed in a single forward pass without generating text.

Reordering multiple-choice options can change a model's answer, an effect studied as selection bias (a preference for certain option letters) and as order sensitivity (changes in answers across option orderings).

When does a language model acquire a capability?

Recent advances in multimodal large language models (MLLMs) have enabled them to act as autonomous agents, creating and editing 3D models via code or direct software manipulation.

Fine-grained visual understanding requires distinguishing subtle differences in local attributes and spatial relations, yet remains challenging for existing vision-language models.

Emerging decision models such as JEV provide typed interfaces that map task instructions and observations directly to choices, judgments, and scores.

AI agents are demonstrating increasingly strong offensive and defensive cybersecurity capabilities in the real world.

Long-horizon agents accumulate progress in persistent task state, yet repeatedly process interaction histories describing how that progress was achieved.

Evaluation for automatic crisis reporting should measure what practitioners need, not what is convenient to compare.

Post-training is crucial for improving text-to-image models, yet existing pipelines typically rely on manually curated training data and fixed training distributions.

We introduce LensErase, a general-purpose mask-based Erase Anything foundation model designed to cover a broad spectrum of real-world erase scenarios.

Effective collaboration among large language model (LLM) agents in multi-agent systems requires sharing relevant evidence, requesting missing information, and incorporating teammates' contributions into decisions.