Curating the frontier of Artificial Intelligence

HarnessAPI: A Skill-First Framework for Unified Streaming APIs and MCP Tools

如今，部署为 LLM 工具的每个 Python 函数都必须以两种形式存在: 用于面向人类的客户端和 CI 管道的 HTTP 端点, 和一个 MCP...

Every Python function deployed as an LLM tool must today exist in two forms: an HTTP endpoint for human-facing clients and CI pipelines, and an MCP to...

Towards a General Intelligence and Interface for Wearable Health Data

虽然无处不在的可穿戴传感器捕获了大量的行为和生理信息,，但有效地将这些信号转化为个性化...

While ubiquitous wearable sensors capture a wealth of behavioral and physiological information, effectively transforming these signals into personaliz...

Advancing Mathematics Research with AI-Driven Formal Proof Search

大型语言模型 (LLMs) 越来越擅长数学推理,，但它们的不可靠性限制了它们在数学研究中的实用性。一个米特...

Large language models (LLMs) increasingly excel at mathematical reasoning, but their unreliability limits their utility in mathematics research. A mit...

Deep Reinforcement Learning for Flexible Job Shop Scheduling with Random Job Arrivals

灵活作业车间调度问题 (FJSP) 是一组作业到机器的最佳分配。 FJSP: 联合国仍然存在两个主要挑战...

The Flexible Job Shop Scheduling Problem (FJSP) is the optimal allocation of a set of jobs to machines. Two primary challenges persist in FJSP: the un...

LCGuard: Latent Communication Guard for Safe KV Sharing in Multi-Agent Systems

基于大型语言模型 (LLM) 的多代理系统越来越依赖中间通信来协调复杂的任务。虽然大多数现有...

Large language model (LLM)-based multi-agent systems increasingly rely on intermediate communication to coordinate complex tasks. While most existing ...

Gated DeltaNet-2: Decoupling Erase and Write in Linear Attention

线性注意力用固定大小的循环状态,取代了softmax注意力的无界缓存，将序列混合减少到线性时间和解码...

Linear attention replaces the unbounded cache of softmax attention with a fixed-size recurrent state, reducing sequence mixing to linear time and deco...

MOSS: Self-Evolution through Source-Level Rewriting in Autonomous Agent Systems

自主代理系统在部署后基本上是静态的:，它们不会从用户交互中学习,，并且反复出现的故障持续存在，直到下一个......

Autonomous agentic systems are largely static after deployment: they do not learn from user interactions, and recurring failures persist until the nex...

PALS: Power-Aware LLM Serving for Mixture-of-Experts Models

大型语言模型 (LLM) 推理已成为现代数据中心的主要工作负载, 显着推动 GPU 利用率和能耗...

Large language model (LLM) inference has become a dominant workload in modern data centers, driving significant GPU utilization and energy consumption...

Mind the Sim-to-Real Gap & Think Like a Scientist

假设规划者有一个针对顺序决策问题的预训练模拟器，并且可以选择在现场运行真实实验。该模拟器是...

Suppose a planner has a pre-trained simulator of a sequential decision problem and the option to run real experiments in the field. The simulator is c...

AiraXiv: An AI-Driven Open-Access Platform for Human and AI Scientists

人工智能 (AI) 的最新进展加速了人类创作和人工智能生成的研究成果的增长, 增加了...

Recent advances in artificial intelligence (AI) have accelerated the growth of both human-authored and AI-generated research outputs, placing increasi...