Deploying Large Language Model (LLM) services at the edge benefits latency-sensitive and privacy-aware applications. However, the stateless nature of LLMs makes managing user context (e.g., sessions, preferences) across geo-distributed edge nodes challenging. Existing solutions, such as client-side context storage, often introduce network latency and bandwidth overhead, undermining the advantages of edge deployment. We propose DisCEdge, a distributed context management system that stores and replicates user context in tokenized form across edge nodes. By maintaining context as token sequences rather than raw text, our system avoids redundant computation and enables efficient data replication. We implement and evaluate an open-source prototype in a realistic edge environment with commodity hardware. We show DisCEdge improves median response times by up to 14.46% and lowers median inter-node synchronization overhead by up to 15% compared to a raw-text-based system. It also reduces client request sizes by a median of 90% compared to client-side context management, while guaranteeing data consistency.

The rapid evolution of agentic AI marks a new phase in artificial intelligence, where Large Language Models (LLMs) no longer merely respond but act, reason, and adapt. This survey traces the paradigm shift in building agentic AI: from Pipeline-based systems, where planning, tool use, and memory are orchestrated by external logic, to the emerging Model-native paradigm, where these capabilities are internalized within the model's parameters. We first position Reinforcement Learning (RL) as the algorithmic engine enabling this paradigm shift. By reframing learning from imitating static data to outcome-driven exploration, RL underpins a unified solution of LLM + RL + Task across language, vision and embodied domains. Building on this, the survey systematically reviews how each capability -- Planning, Tool use, and Memory -- has evolved from externally scripted modules to end-to-end learned behaviors. Furthermore, it examines how this paradigm shift has reshaped major agent applications, specifically the Deep Research agent emphasizing long-horizon reasoning and the GUI agent emphasizing embodied interaction. We conclude by discussing the continued internalization of agentic capabilities like Multi-agent collaboration and Reflection, alongside the evolving roles of the system and model layers in future agentic AI. Together, these developments outline a coherent trajectory toward model-native agentic AI as an integrated learning and interaction framework, marking the transition from constructing systems that apply intelligence to developing models that grow intelligence through experience.

Large language model (LLM) agents are fundamentally constrained by context length on long-horizon tasks. We introduce Context-Folding, a framework that empowers agents to actively manage their working context. An agent can procedurally branch into a sub-trajectory to handle a subtask and then fold it upon completion, collapsing the intermediate steps while retaining a concise summary of the outcome. To make this behavior learnable, we develop an end-to-end reinforcement learning framework FoldGRPO with specific process rewards to encourage effective task decomposition and context management. On complex long-horizon tasks (Deep Research and SWE), our folding agent matches or outperforms the ReAct baselines while using an active context 10$\times$ smaller and significantly outperforms models that rely on summarization-based context management.

Long-horizon tasks that require sustained reasoning and multiple tool interactions remain challenging for LLM agents: small errors compound across steps, and even state-of-the-art models often hallucinate or lose coherence. We identify context management as the central bottleneck -- extended histories cause agents to overlook critical evidence or become distracted by irrelevant information, thus failing to replan or reflect from previous mistakes. To address this, we propose COMPASS (Context-Organized Multi-Agent Planning and Strategy System), a lightweight hierarchical framework that separates tactical execution, strategic oversight, and context organization into three specialized components: (1) a Main Agent that performs reasoning and tool use, (2) a Meta-Thinker that monitors progress and issues strategic interventions, and (3) a Context Manager that maintains concise, relevant progress briefs for different reasoning stages. Across three challenging benchmarks -- GAIA, BrowseComp, and Humanity's Last Exam -- COMPASS improves accuracy by up to 20% relative to both single- and multi-agent baselines. We further introduce a test-time scaling extension that elevates performance to match established DeepResearch agents, and a post-training pipeline that delegates context management to smaller models for enhanced efficiency.

Large Language Models (LLMs) are increasingly utilized in multi-agent systems (MAS) to enhance collaborative problem-solving and interactive reasoning. Recent advancements have enabled LLMs to function as autonomous agents capable of understanding complex interactions across multiple topics. However, deploying LLMs in MAS introduces challenges related to context management, response consistency, and scalability, especially when agents must operate under memory limitations and handle noisy inputs. While prior research has explored optimizing context sharing and response latency in LLM-driven MAS, these efforts often focus on either fully centralized or decentralized configurations, each with distinct trade-offs. In this paper, we develop a probabilistic framework to analyze the impact of shared versus separate context configurations on response consistency and response times in LLM-based MAS. We introduce the Response Consistency Index (RCI) as a metric to evaluate the effects of context limitations, noise, and inter-agent dependencies on system performance. Our approach differs from existing research by focusing on the interplay between memory constraints and noise management, providing insights into optimizing scalability and response times in environments with interdependent topics. Through this analysis, we offer a comprehensive understanding of how different configurations impact the efficiency of LLM-driven multi-agent systems, thereby guiding the design of more robust architectures.

Recent advancements in Large Language Model (LLM)-based frameworks have extended their capabilities to complex real-world applications, such as interactive web navigation. These systems, driven by user commands, navigate web browsers to complete tasks through multi-turn dialogues, offering both innovative opportunities and significant challenges. Despite the introduction of benchmarks for conversational web navigation, a detailed understanding of the key contextual components that influence the performance of these agents remains elusive. This study aims to fill this gap by analyzing the various contextual elements crucial to the functioning of web navigation agents. We investigate the optimization of context management, focusing on the influence of interaction history and web page representation. Our work highlights improved agent performance across out-of-distribution scenarios, including unseen websites, categories, and geographic locations through effective context management. These findings provide insights into the design and optimization of LLM-based agents, enabling more accurate and effective web navigation in real-world applications.

In this paper, we make a case for a proxy for large language models which has explicit support for cost-saving optimizations. We design LLMProxy, which supports three key optimizations: model selection, context management, and caching. These optimizations present tradeoffs in terms of cost, inference time, and response quality, which applications can navigate through our high level, bidirectional interface. As a case study, we implement a WhatsApp-based Q&A service that uses LLMProxy to provide a rich set of features to the users. This service is deployed on a small scale (100+ users) leveraging the cloud; it has been operational for 15+ weeks and users have asked 1400+ questions so far. We report on the experiences of running this service as well as microbenchmark the specific benefits of the various cost-optimizations we present in this paper.

Understanding the direction and context of an ever-evolving conversation is beneficial to building natural, human-like conversational interfaces. Being able to classify utterances as the conversation develops requires managing context across multiple turns. Consider a caller who asks their financial planner for insights regarding their monthly expenses: "What were my expenses this year?" They may also ask for more granular information, such as "How about for last month?" As the conversation progresses, the bot needs to understand if the context is changing and adjust its responses accordingly.