What makes great scientists great? Even as AI systems start to make progress on open problems, scientists remain far ahe…

LLM assistants are widely used for daily social advice, yet evaluating their social reasoning in such consultation setti…

Complex documents such as scientific papers contain rich information in tables, figures, and equations; however, existin…

We define OmniVChat (Omni Video Chat) as the task of native audio-visual dialogue between a user and an omni model. In O…

APort Vault is a benchmark for payment authorization in tool-using AI agents. It replays 4,371 attacks written by humans…

The widespread adoption of long-horizon agents has made model workloads increasingly input-heavy. Although prior work ha…

As corporate AI adoption continues to grow, enterprise-grade LLM agents are being deployed into sensitive contexts such …

Fine-tuning instruct models often improves target performance while inducing behavioral drift from the reference model, …

Concurrent local LLM serving on unified-memory desktops must preserve memory headroom and output fidelity, which speed-o…

Benchmark researchers and developers of large language models (LLMs) and other AI systems need to find relevant evaluati…

Research taste, the judgment of what to read, which problems to pursue, and which methods to try, increasingly shapes th…

Computer graphics is built on code. Large language model (LLM) agents can now generate graphics code and iteratively ref…

As publications and products on time series forecasting (TSF) proliferate, researchers face a growing burden of analyzin…

Do research agents have taste, the judgment of which experiments are worth running before their outcomes are known? We g…

Reproducing a published 3DGS paper takes weeks of expert engineering before any extension can be proposed. Prior autonom…

Recent AI systems have made rapid progress in scientific discovery when given well-defined metrics, but whether they can…

AI reviewers can assign different judgments to manuscripts that report the same science in different wording, potentiall…

Difference-in-differences (DID) studies are widely used to evaluate climate policy, but assessing the evidence supportin…

Mixture-of-Experts (MoE) scales capacity, but existing designs cannot set three quantities independently. For a single t…

Professional graphic design is a long-horizon agentic task in which structured, editable artifacts emerge from many inte…

A pretrained robot foundation policy may execute most of a long-horizon task yet repeatedly fail at a few critical subta…

Training capable coding agents via reinforcement learning (RL) requires diverse tasks with reliable verifiers. Open-sour…

Reference-to-video (R2V) generation is evolving toward increasingly general and versatile reference control, giving rise…

We present MintAct, a family of vision-language models that unifies UI grounding, multi-step navigation across mobile, d…

On-policy distillation (OPD) improves reasoning models by learning the token-level discrepancy between a stronger teache…

Skills can improve the performance of Large Language Model (LLM) agents by providing task-specific procedural guidance, …

Computer-use agents (CUAs) have advanced along two separate lines: graphical interaction and software development throug…

We need studies on conversational AI (CAI) at scale to understand human behavior and shape CAI design. However, fragment…

Question answering has advanced rapidly with large language models, but predominantly for high-resource languages, in bo…

Text-guided image editing must introduce the requested changes while preserving unrelated source content. Diffusion-base…

Paper2Agent, published in Nature, converts papers into validated MCP tools, scoring 91.2% on 300 questions across 74 pap…

Document parsing converts document images into structured content and requires reliable performance across diverse layou…

Telecom fraud scripts evolve rapidly and are often designed to resemble routine service conversations, creating two key …

Multi-turn agents trained with reinforcement learning (RL) receive a single scalar reward per trajectory, which motivate…

On-policy self-distillation (OPSD) lets a language model learn from a frozen copy of itself that sees an answer or a wor…

Large Language Models (LLMs) are increasingly deployed in applications that must weigh clashing moral values, yet even s…

We study length inflation in on-policy distillation (OPD), where student responses can become excessively long and even …

Creating deformable assets for robot manipulation requires jointly specifying their geometry, appearance, and physical p…

Modeling articulated objects from sparse monocular views is challenging because each observation reveals only partial ge…

As coding agents move from supervised code completion to unattended, around-the-clock exploration, their work expands fr…

Agent trajectories record what an agent does and what happens next. Yet standard supervised fine-tuning (SFT) applies lo…

Information retrieval is increasingly important as LLM agents tackle complex tasks involving diverse information needs. …

Professional design requires any-color control: the ability to specify an object's target color with any 24-bit hex valu…

Large Reasoning Models (LRMs) achieve strong performance on complex tasks but exhibit systematic inefficiency: they ofte…

Multi-modal image generation, particularly subject-driven customization, has garnered growing attention in recent years.…

Coding harnesses shape how autonomous coding agents translate model capabilities into long-horizon software-engineering …

Spatial intelligence requires more than describing object locations. Under incomplete observation, models must identify …

Induced pluripotent stem cell (iPSC) culture increasingly relies on segmentation foundation models, yet deployment on la…

Video diffusion models repeatedly process long spatiotemporal token sequences during denoising, making attention a major…

3D diffusion policies are strong at generating geometrically grounded actions from current observations, but successful …