Machine Learning System

Despite their remarkable success in complex tasks propelling widespread adoption, large language-model-based agents still face critical deployment challenges due to prohibitive latency and inference costs. While recent work has explored various methods to accelerate inference, existing approaches suffer from significant limitations – they either fail to preserve performance fidelity, require extensive offline training of router modules, or incur excessive operational costs. Moreover, they provide minimal user control over the tradeoff between acceleration and other performance metrics. To address these gaps, we introduce Dynamic Speculative Planning (DSP), an asynchronous online reinforcement learning framework that provides lossless acceleration with substantially reduced costs without requiring additional pre-deployment preparation. DSP explicitly optimizes a joint objective balancing end-to-end latency against dollar cost, allowing practitioners to adjust a single parameter that steers the system toward faster responses, cheaper operation, or any point along this continuum. Experiments on two standard agent benchmarks demonstrate that DSP achieves comparable efficiency to the fastest lossless acceleration method while reducing total cost by 30% and unnecessary cost up to 60%.

Conventional operating system scheduling algorithms are largely content-ignorant, making decisions based on factors such as latency or fairness without considering the actual intents or semantics of processes. Consequently, these algorithms often do not prioritize tasks that require urgent attention or carry higher importance, such as in emergency management scenarios. However, recent advances in language models enable semantic analysis of processes, allowing for more intelligent and context-aware scheduling decisions. In this paper, we introduce the concept of semantic scheduling in scheduling of requests from large language models (LLM), where the semantics of the process guide the scheduling priorities. We present a novel scheduling algorithm with optimal time complexity, designed to minimize the overall waiting time in LLM-based prompt scheduling. To illustrate its effectiveness, we present a medical emergency management application, underscoring the potential benefits of semantic scheduling for critical, time-sensitive tasks.

Agents, as user-centric tools, are increasingly deployed for human task delegation, assisting with a broad spectrum of requests by generating thoughts, engaging with user proxies, and producing action plans. However, agents based on large language models (LLMs) often face substantial planning latency due to two primary factors – the efficiency limitations of the underlying LLMs due to their large size and high demand, and the structural complexity of the agents due to the extensive generation of intermediate thoughts to produce the final output. Given that inefficiency in service provision can undermine the value of automation for users, this paper presents a human-centered efficient agent planning method – Interactive Speculative Planning – aiming at enhancing the efficiency of agent planning through both system design and human-AI interaction. Our approach advocates for the co-design of the agent system and user interface, underscoring the importance of an agent system that can fluidly manage user interactions and interruptions. By integrating human interruptions as a fundamental component of the system, we not only make it more user-centric but also expedite the entire process by leveraging human-in-the-loop interactions to provide accurate intermediate steps. Code and data will be released.

This paper envisions a revolutionary AIOS-Agent ecosystem, where Large Language Model (LLM) serves as the (Artificial) Intelligent Operating System (IOS, or AIOS)–an operating system “with soul”. Upon this foundation, a diverse range of LLM-based AI Agent Applications (Agents, or AAPs) are developed, enriching the AIOS-Agent ecosystem and signaling a paradigm shift from the traditional OS-APP ecosystem. We envision that LLMs impact will not be limited to the AI application level, instead, it will in turn revolutionize the design and implementation of computer system, architecture, software, and programming language, featured by several main concepts. LLM as OS (system-level), Agents as Applications (application-level), Natural Language as Programming Interface (user-level), and Tools as Devices/Libraries (hardware/middleware-level). In this paper, we begin by introducing the architecture and historical evolution of traditional Operating Systems (OS). Then we formalize a conceptual framework for AIOS through “LLM as OS (LLMAO)”, drawing analogies between AIOS components and traditional OS elements. LLM is likened to OS kernel, context window to memory, external storage to file system, hardware tools to peripheral devices, software tools to programming libraries, and user prompts to user commands. Subsequently, we introduce the new AIOS-Agent Ecosystem, where users and developers can easily program Agent Applications (AAPs) using natural language, democratizing the development of and the access to computer software, which is different from the traditional OS-APP ecosystem, where desktop or mobile applications (APPs) have to be programmed by well-trained software developers using professional programming languages. Following this, we explore the diverse scope of Agent Applications. These agents can autonomously perform diverse tasks, showcasing intelligent task-solving ability in various scenarios. We delve into both single agent systems and multi-agent systems, as well as human-agent interaction. Lastly, we posit that the AIOS-Agent ecosystem can gain invaluable insights from the development trajectory of the traditional OS-APP ecosystem. Drawing on these insights, we propose a strategic roadmap for the evolution of the AIOS-Agent ecosystem. This roadmap is designed to guide the future research and development, suggesting systematic progresses of AIOS and its Agent applications.