Long-Horizon Agents

Agentic systems have rapidly advanced in their ability to interact with real-world environments, leverage external tools, and provide services for users. However, unlike natural-world tasks that assume well-defined instructions, human-centered scenarios are characterized by ambiguous requests that lead to large, open-ended solution spaces. Decoding users' personalized preferences is therefore essential for narrowing the candidate solution space. This introduces a new challenge, personalized agentic reasoning, which requires agents to jointly interact with both users and environments to deliver personalized services. In this paper, we present ODYSSE, a Reinforced Fine-Tuning (RFT) framework for personalized agentic reasoning. At its core, ODYSSE proposes Episode-wise GRPO (ESPO), a novel extension of Group Relative Policy Optimization (GRPO) designed to address long action horizons and strong cross-step dependencies in personalized agentic reasoning. Rather than optimizing individual steps independently, ESPO introduces an episode-level reward mechanism together with episodic advantage estimation, enabling upstream evidence to effectively guide downstream personalized decisions and allowing agents to progressively resolve ambiguous user requests across multiple interaction steps. We further propose an episodic batch sampler that groups actions from the same episode into unified training batches, facilitating coherent optimization under ESPO. We evaluate ODYSSE on realistic long-horizon personalized GUI reasoning tasks. Experimental results demonstrate that ODYSSE consistently outperforms both specialist and general-purpose LVLMs, highlighting its effectiveness for personalized agentic reasoning.

Machine learning engineering (MLE) tasks require long-horizon decision making over iterative solution debugging and refinement, under expensive and feedback-driven environment interactions. Developing and training a monolithic agent for such tasks is fundamentally challenging, as it must simultaneously manage extremely long and noisy contexts, explore vast solution spaces, and remain effective under limited model capacity and computational budgets. To address these challenges, we propose Matryoshka Agent, a unified hierarchical agent framework for complex long-horizon tasks. Matryoshka Agent decomposes agentic problem solving into a coordinated hierarchy of decision making and execution: a high-level Orchestrator maintains compact, long-horizon exploration states and issues strategic instructions, while lower-level Sub-Agents execute concrete solution attempts through direct environment interaction, mediated by standardized Tool interface. This design decouples strategic exploration from costly execution, substantially reducing the burden of long-context reasoning and enabling efficient iterative refinement. We further develop an efficient training paradigm for Matryoshka Agent. Experimental results on a broad range of MLE tasks with diverse model types and scales demonstrate that Matryoshka Agent is an effective and scalable paradigm for long-horizon MLE tasks and complex agentic problem solving. Notably, Matryoshka Agent enables Qwen3-4B-Instruct to reach Orchestrator performance comparable to o4-mini. Applying Matryoshka Agent to Qwen3-30B-Coder results in at most 36.7% relative performance gain.

Long-horizon LLM agents accumulate reasoning traces, actions, and tool observations that can eventually exceed a model's fixed context window. Existing compaction methods address this limitation by discarding, summarizing, or retrieving earlier information, but they may remove task-critical details or fail to recover them reliably. We propose ARC (Addressable Recall Compaction), a context-management framework that separates archival storage from active-context presentation. ARC stores tool observations in an append-only, ID-addressable log and replaces older observations with compact citations when compaction is required. The agent can subsequently use these identifiers to request stored content without re-executing the corresponding tools or depending solely on similarity-based retrieval. We evaluate ARC using Qwen3-8B with a 16k context window and Qwen3-32B with a 32k context window. On the Needle-in-a-Haystack evaluation, ARC achieves an average exact-answer accuracy of 99.40%, compared with 88.12% for the best-performing baseline in our evaluation. ARC also reduces estimated serving time and HBM traffic under our hardware-cost model. On the LongBench-v2 Hard subset, ARC obtains an average accuracy of 29.97%, compared with 28.25% for the best-performing baseline. These results indicate that explicit, address-based recall can improve information retention and serving efficiency relative to the evaluated context-management baselines under the tested settings.

Agentic tasks are inherently long-horizon and multi-turn, constantly accumulating context through interactions with the environment. Existing context compression methods inevitably incur information loss and are triggered by rigid heuristic rules, leaving them misaligned with the agent's evolving reasoning focus. We propose Agentic Context Management (ACM), a framework that equips agents with purpose-built context editing tools for lossless context management. Inspired by the interaction between short-term and long-term human memory, the agent autonomously decides when to compress its context, offloads discarded content to an external memory system, and queries it on demand for later retrieval. Building on this framework, we further develop a post-training pipeline that constructs high-quality demonstrations of context management and improves model performance on both agentic search and coding tasks. Further analysis reveals that effective context management reduces peak token pressure, enables extended explorations, and yields more consistent solutions across independent trials. Code, data, and model checkpoints are available at https://github.com/lixiaochuan2020/agentic-context-management.

Recent advances in large language models (LLMs) have enabled search agents to autonomously tackle complex tasks across extended search and reasoning horizons. However, training effective search agents remains challenging due to the lack of scalable and long-horizon tasks, and the difficulty of evaluating and correcting intermediate reasoning and tool-use behaviors. We introduce SearchArt, a scalable framework for training long-horizon search agents through verification-driven task synthesis and a multi-stage post-training pipeline. SearchArt constructs large-scale datasets for complex search-, research- and user-oriented tasks by synthesizing diverse information-seeking QA pairs and corresponding search trajectories from web documents and automatically generated evidence graphs. To ensure the reliability of the synthesized data, we design a verification pipeline that jointly evaluates QA consistency, trajectory quality, and the relevance of retrieved evidence. The verified trajectories are subsequently used in a multi-stage training process comprising supervised fine-tuning and reinforcement learning-based policy optimization. Search agents trained with SearchArt exhibit adaptive search planning, iterative evidence aggregation, and complex reasoning over extended interaction horizons. Experimental results demonstrate that, with only (Qwen3.5-) 27B parameters, SearchArt scores 74.39 on BrowseComp-ZH, 70.06 on BrowseComp, and 52.55 on Deepresearch-bench, matching or surpassing frontier closed-source agents on both deepsearch and deepresearch benchmarks.

Long-horizon tasks require sustained perception, reasoning, and exploration, and are a persistent challenge for large language model (LLM) agents. This gap is reflected in their limited performance on continual learning benchmarks such as ARC-AGI-3, especially when models are evaluated out of the box. Various agent harnesses have been proposed to close this gap, and each commits to a strategy for handling long sequences of observations, i.e., what information to save from the environment and how to load it into model context, a choice we argue is particularly consequential. Existing methods for context management face a significant tradeoff, as preserving more information makes retrieving relevant details less tractable. We propose PRO-LONG, a minimal context management framework built around programmatic memory for LLM agents in long-horizon, exploratory settings. PRO-LONG addresses the tradeoff by keeping a complete, structured interaction log and capitalizing on recent progress in coding agents to search this history efficiently. On the full ARC-AGI-3 public game set, PRO-LONG improves over a base coding agent by an average of 18.0 percentage points across frontier models, and matches or exceeds state-of-the-art specialized harnesses (up to 76.1% pass@1) while using 4.2-5.8x fewer tokens. With Fable 5, PRO-LONG achieves 97.4% best@2 at a total cost of $1,750. Relevant code and logs are available at https://github.com/alexisfox7/PRO-LONG.

We present Solar Open 2, a 250B-A15B Mixture-of-Experts language model built for long-horizon agentic tasks, scaled up from Solar Open 1 (Solar Open 100B). To hold entire agent trajectories in a single context, Solar Open 2 reaches a 1M-token window through a hybrid attention stack that interleaves one softmax layer among every three linear-attention layers, using no positional encoding and a gated delta rule extended to negative eigenvalues. To train at this scale under a fixed compute budget, we make training efficient in two ways: a stronger starting point, and higher-value data. For the starting point, we initialize Solar Open 2 from Solar Open 1, transferring the 5.69B-parameter shared skeleton that survives the architectural change and learning everything else through full pre-training. For the data, we curate for value per token: quality- and rarity-aware data curation and mixture-ratio optimization refine a 20T pool into a 10T mixture that, at equal token budget, outperforms the Solar Open 1 recipe. To build its agent skills, we train twelve domain specialists across purpose-built scenarios, then consolidate them into a single model by Multi-teacher On-Policy Distillation (MOPD). Against comparably sized open-weight models on English benchmarks, Solar Open 2 leads on MMLU-Pro, LiveCodeBench, and the APEX-Agents agentic suite, and stays competitive with the strongest (DeepSeek-V4-Flash and MiMo-V2.5) elsewhere. On Korean benchmarks, Solar Open 2 records the highest average of any model compared, including fast-tier closed APIs, and on Ko-GDPval, an in-house Korean officework-agent benchmark, it is competitive with DeepSeek-V4-Pro (1.6T) at less than a sixth of its size.

Group-based policy optimization has been increasingly used to train large language model (LLM) agents from sparse outcome rewards by comparing trajectories or steps within a group. However, on difficult long-horizon tasks, this comparison can suffer from a sampling imbalance: repeated or low-effect actions dominate the high-probability region of the policy while useful state-changing actions remain under-sampled. This imbalance produces many all-failed rollout groups, where outcome rewards provide no direction for correcting the policy. Together, these effects can form a self-reinforcing credit trap: failure-dominated sampling yields no outcome-based correction, allowing repeated low-effect actions to persist. To break this loop, we propose Progress-conditioned Group Policy Optimization (ProGPO), which uses first-visit observation coverage only when all samples in a group receive zero outcome reward. Specifically, within such groups, ProGPO assigns higher relative advantages to trajectories or steps that visit more new states since reaching new observations is a prerequisite for task success. Experiments on two challenging agentic benchmarks, ALFWorld and WebShop with Qwen2.5-1.5/7B-Instruct, show that ProGPO consistently improves over group-based baselines, with particularly large gains on hard tasks.

Long-horizon rollout generation has become the dominant systems bottleneck in agentic reinforcement learning (RL). As agents interact with environments over many turns, trajectories rapidly grow to tens of thousands of tokens, making synchronous RL training increasingly constrained by rollout. We propose WAR, a workload-aware rollout system that substantially accelerates synchronous agentic RL by jointly optimizing decoding and scheduling. WAR is built on a key observation: the optimal rollout optimization strategy depends on runtime load: (1) Under low load, WAR enables model-free speculative decoding with SuffixDecoding, which reuses suffix patterns from previously completed trajectories as speculative drafts for future rollouts. Unlike model-based drafters, SuffixDecoding introduces no additional draft model and avoids GPU contention with rollout generation. (2) Under high load, where saturated batched decoding leaves limited room for speculative speedup, WAR shifts the optimization focus to cache-aware scheduling. A global scheduler places requests across rollout replicas based on cache locality, trajectory progress and server load, reducing redundant KV-cache recomputation and mitigating load imbalance. By combining decoding-level suffix reuse with system-level rollout scheduling, WAR delivers robust throughput improvements across workload regimes without changing the underlying RL algorithm. WAR improves long-context agentic rollout throughput by 1.4x under low load and up to 1.6x under high load. These results show that WAR removes a major rollout bottleneck in synchronous agentic RL and provides a practical path toward scalable long-context agent training.

Existing memory systems for long-horizon LLM agents often retrieve prior traces as passive context rather than converting them into executable capabilities. In this paper, we propose MSCE, a training-free Memory--Skill Co-Evolution framework that organizes agent experience into grounded step traces, reusable procedural policies, and declarative environmental cognition. MSCE crystallizes evidence-backed L2 policies with positive estimated gain into callable skills that retain evidence links, applicability boundaries, decision guidance, verification rules, and reliability estimates. It further introduces reflection-weighted value backfilling, which propagates sparse terminal feedback through dense local self-reflections to produce evidence-calibrated trace values for governing memory and skill evolution. Experiments on EvoAgentBench and LoCoMo demonstrate that MSCE significantly outperforms state-of-the-art skill-augmented and memory-driven agent baselines, exhibiting strong cross-domain transferability and lifelong-evolution capabilities.

Long-horizon AI agents are becoming increasingly capable, yet their interaction with users remains surprisingly thin. In most workflows, users give an initial instruction, receive only selective textual updates, and lose a clear sense of what the agent is doing or when to step in. This leaves a missing part in the current agent ecosystem: an always-on Jarvis-style mediator that keeps the agent continuously reachable to the user. Such a mediator should support real-time spoken interaction with the user, answer questions without interrupting the worker, proactively report progress or confusion, and inject user guidance back into the agent's execution when useful. In this work, we introduce JarvisBench, a benchmark for measuring the dual value of mediation in long-horizon agent workflows. JarvisBench contains two complementary tracks: an agent-collaboration track that measures whether mediation improves downstream task completion, and a user-interaction track that measures whether mediation makes ongoing execution more understandable, responsive, and accessible to users. We instantiate the benchmark with a modular reference Jarvis prototype and evaluate it on 34 text-only WildClaw tasks executed in OpenClaw. Preliminary results with GPT-5.5, Claude Opus 4.7, Gemini-based, and GPT-based worker agents suggest that Jarvis-style mediation can provide trace-grounded responses to user questions and improve task performance when sparse user guidance is injected at appropriate moments. The results also show that effectiveness depends strongly on the mediator's LLM brain, highlighting both the promise of this missing middle layer and the need for broader community effort. Demo page https://cchen1436.github.io/jarvis

While LLM agents demonstrate strong reasoning abilities in compact and well-defined scenarios, they struggle to maintain robustness and effectiveness when faced with large-scale, diverse, and dynamic real-world environments that demand seamless tool integration. To address this gap, we introduce ToolVerse, a comprehensive framework that scales up agentic RL environments and enables agents to perform complex long-horizon reasoning in Tool-Integrated Reasoning (TIR) tasks. First, ToolVerse automatically builds the massive executable agent training environments from nearly 400 real-world Model Context Protocols (MCPs) that contain about 4500 tools. Second, we propose a task design strategy based on a tool dependency graph, utilizing Dynamic Unlocking Sampling Algorithm to generate long-horizon tasks, and produce GUST (Graph Unlocking Sampling Tasks) dataset. Third, to alleviate the credit assigment problem in long-horizon agentic RL, we propose a fine-grained Turn-Aware Relative Advantage algorithm. We conduct extensive Agentic RL training using ToolVerse and evaluate our framework on serveral agentic benchmarks. Experimental results demonstrate that our framework significantly strengthens LLMs' capabilities in long-horizon tool use, achieving a marked performance boost and showcasing robust reasoning within dynamic environments.

Large language model (LLM)-based generative agents simulate human behavior through long-horizon decision-making processes that comprise intermediate steps such as planning, memory retrieval, reflection, and action selection. However, fine-grained human annotations of these intermediate steps remain scarce, and existing agents are not grounded in human preferences over such intermediate decisions. To address this gap, we introduce \method, an interactive simulation interface that enables us to collect step-level human preference supervision over agent decision trajectories, leading to a dataset of 57K fine-grained annotations. We conduct step-level preference learning on open-weight language models using supervised finetuning and direct preference optimization on this data, consistently improving simulation fidelity, coordination, and interaction quality, and inducing more socially effective agent behavior. Our results show that step-level human supervision is an effective training signal for improving both local decision quality and long-horizon agent behavior.

Multi-turn agents solve complex tasks through extended sequences of tool interactions before producing a final answer, making credit assignment a fundamental challenge during post-training. Outcome rewards provide reliable supervision for short-horizon reasoning, but become sparse and high-variance as trajectories grow to tens or hundreds of tool calls. They can also be misleading: a failed rollout may contain many useful actions that move the agent closer to the goal, yet outcome-only training assigns them the same negative advantage as the eventual mistake. We propose TRACE (Turn-level Reward Assignment via Credit Estimation), a dense credit-assignment method for agentic reinforcement learning. TRACE represents rollouts as state transitions at tool-call boundaries, obtains gold-answer log-probabilities from a frozen reference model, transforms them into log-ratio state values, and derives per-action rewards as Temporal-Difference changes in those values. This requires no additional critic or process-label training, and its one-step log-ratio TD component telescopes across redundant tool calls. On long-horizon complex search, TRACE substantially improves base-model tool-use ability using pure RL, without a cold-start supervised fine-tuning stage, an agentic mid-training stage, or training on live-web data. On the closed-web BrowseComp-Plus benchmark, it raises Qwen3-4B from $7.2$ to $35.6$ and Qwen3-30B-A3B from $8.4$ to $42.6$. The learned search behavior also transfers to open-web benchmarks, and the learning curves show earlier improvement and faster convergence during RL training.

Agent memory is a systems problem for long-horizon agents. Practical deployments require retention of task state across extended conversations, recovery of user-specific facts and preferences across sessions, and accumulation of procedural knowledge from prior outcomes. These requirements extend beyond document retrieval: a memory layer must determine which interactions become durable state, how that state is scoped, how it is retrieved under latency constraints, and how it is revised or removed over time. This report studies Oracle Agent Memory as a database-native memory substrate built on Oracle Database. Three themes organize the discussion: memory as a lifecycle spanning ingestion, extraction, consolidation, retrieval, summarization, and revision or removal; a layered architecture that separates an active memory core from a passive memory-store interface with explicit scope control across users, agents, and threads; and evaluation methodology in which downstream task accuracy is complemented by memory-centric measures such as evidence retrieval, recall, latency, and estimated token use. The report summarizes LongMemEval results, reaching 93.8% accuracy, compares Oracle Agent Memory against flat-history baselines, using about 10.7x fewer tokens, and published or reported external baselines where available, and closes with implementation-oriented appendix material covering setup, thread lifecycle, and search semantics.

Recent advances in large language models (LLMs) and vision-language models (VLMs) have enabled increasingly capable digital agents for computer use. However, real-world tasks are often long-horizon and involve evolving contexts containing accumulated observations, intermediate edits, failed attempts, and partially completed executions. Existing agents typically operate over raw interaction history, making task progress difficult to interpret, verify, and recover, which ultimately limits reliable long-horizon execution. In this paper, we argue that addressing this challenge requires explicitly structuring both the agent's state and workflow around a unified causal representation of task progress. We present \textbf{StructAgent}, a state-centered framework that introduces a unified state for maintaining compact, verifiable task progress and a structured workflow that regulates progress through verifier-backed state transitions. Building on this design, StructAgent further enables explicit progress checkpointing, evidence-driven task completion, targeted failure recovery, and tool-supported execution, while ensuring that all progress updates remain grounded in verification. Extensive experiments demonstrate that StructAgent consistently improves a wide range of LLM and VLM backbones on long-horizon computer-use tasks. On OSWorld-Verified, it improves Qwen3.5-9B from 27.0% to 46.9% success rate and Qwen3.5-27B from 31.6% to 62.2%, while achieving a new open-source state of the art of 78.9% with MiniMax-M3. Moreover, the same framework generalizes beyond desktop environments to Minecraft, demonstrating the generality of our design.

Long-term physical coexistence with intelligent robots requires more than capable robot policies. A persistent robotic assistant must support diverse user-facing interfaces, maintain long-horizon memory of people and preferences, coordinate across robot embodiments, and translate human intent into safe physical execution. We introduce PHILIA, a multi-robot agent built around a robot gateway abstraction. PHILIA retains the rich interaction and tool ecosystem of OpenClaw while exposing robot-local runtimes, onboard perception, navigation, speaker, and robot policies through a unified capability interface. This design decouples low-frequency, high-semantic agent reasoning from high-frequency, low-level robot execution, enabling plug-and-play integration of user interfaces, robot embodiments, and policy backends. As a result, the user experience becomes compositional: advances in user interfaces, robot embodiments, robot policies, navigation, or interaction algorithms can improve the overall experience without redesigning the system. We validate the architecture on Astribot S1 robots while designing the robot gateway contract to support future heterogeneous robot platforms through a shared capability interface for observation, task execution, navigation, speech playback, status monitoring, and task cancellation. We present representative use cases in which agent memory and scene understanding are grounded in robot actions. These span interactive household scenarios, ranging from simple organization to challenging long-horizon and dexterous service tasks, such as packing a backpack and lifting a garbage bag. We highlight the human-robot interaction flow, where contextual understanding of user intent and preferences, together with human-in-the-loop confirmation or adjustment during execution, is essential for effective assistance.

Memory is becoming a core component of long-horizon AI agents, allowing agents to reuse past experience when operating web browsers, software tools, and other interactive environments. Existing work mostly treats memory as a supply problem, asking what experience to write, how to store it, and which entry to retrieve for the next task. Yet we still lack a clear account of how models consume retrieved memory across a multi-step action trajectory. This consumption process matters because it determines not only what memories should be retrieved, but also what models and control policies are needed to use them safely. To diagnose this process, we propose Entry--Propagation--Recovery (E-P-R), a trajectory-level framework that asks where memory first changes an action, whether that change carries forward, and whether the agent can recover after leaving a correct path. We instantiate E-P-R on WebArena and on MemTrapBench, a controlled benchmark we build to isolate these phases. We find that the main failure often begins at entry: agents adopt conflicting memory at the first exposed decision point even when it is task-wrong. Repeated exposure then amplifies this early error, while recovery after divergence is weak. Together, these effects create a compliance trap: across models, conflicting memory induces similar compliance rates, but once agents comply, their success rates collapse to a low floor. Stronger agents therefore suffer larger absolute damage because each compliance event erases more baseline capability. These results suggest that memory-augmented agents should be evaluated not only by retrieval quality or final success rate, but by how they consume memory throughout the trajectory.

In this work, we introduce LongMedBench, a real-world EHR-based benchmark for long-horizon clinical decision-making. Prior evaluations of LLM-based medical agents have largely emphasized short-context knowledge QA and tool use. However, real-world medical care is inherently longitudinal, and clinicians must aggregate evidence across repeated visits, tests, and evolving treatments. Therefore, long-horizon interaction is essential for realistic assessment. LongMedBench is constructed via a reproducible pipeline that integrates MIMIC-IV admission records and clinical notes into time-series event streams and long-context memory datasets, enabling long-horizon, multi-session interactions between agents and a clinical environment. It comprises 335 patients, with 19.72 inpatient visits per patient on average and 44.91 medical events per visit. Guided by the long-horizon decision process, we propose an evaluation taxonomy with three suites: fact-based QA, temporal reasoning, and long-horizon decision-making. This taxonomy measures how agents understand and leverage historical patient information over extended horizons. Our experiments show that while recent LLMs can make good use of explicit timestamps, they have challenges in implicit time inference; The RAG and agent memory system can improve the performance of information retrieval tasks, but the performance of decision-making tasks is highly dependent on the model's immediate context.

Deployed LLM agents rely on agentic context, the model-external textual control content assembled by an operational harness. In this work, the mutable component of that context is a persistent system-level instruction that is updated from operational experience while the model, tools, and harness remain fixed. Over long evolution horizons, flat-text maintenance makes verification increasingly difficult as accumulated instructions grow and interact. We propose Graph-Regularized Agentic Context Evolution (GRACE), which maintains the persistent instruction component as a typed semantic graph and validates proposed updates within the local typed neighborhoods of modified nodes. Accepted graph updates are reconstructed as incremental edits to the textual instruction checkpoint used at deployment. We evaluate GRACE within a fixed telecom agent harness derived from $τ^2$-bench under a controlled distribution-shift protocol. Across five independent replications, GRACE improves strict reliability, measured by pass^3, from the Gemini 2.5 Flash zero-shot value of 0.091 to 0.673$\pm$0.136 at the final checkpoint. This exceeds a Gemini 3.1 Pro zero-shot reference of 0.242 on the same held-out set, while the flat-text HCE baseline finishes at 0.191$\pm$0.051. These results identify two requirements for reliable long-horizon context evolution, a structural substrate that makes verification local and a consolidation mechanism that keeps accumulated instruction content usable.

AI agents have become capable of autonomously completing short, well-specified tasks. However, existing terminal benchmarks largely focus on simple problems that finish within minutes and are evaluated only by their final outcome. This setup overlooks intermediate progress and partial solutions, yielding sparse reward signals and an incomplete picture of agent capability. We introduce Long-Horizon-Terminal-Bench, a terminal benchmark of 46 long-horizon tasks spanning nine categories, including experiment reproduction, software engineering, multimodal analysis, interactive games, and scientific computing. Each task follows a Terminal-Bench-style setup with a reference solution or simulation engine, but is further decomposed into fine-grained graded subtasks. This design enables dense intermediate rewards and partial credit, allowing evaluation to capture not only whether an agent reaches the final goal, but also how far it progresses on open-ended workflows. Tasks in Long-Horizon-Terminal-Bench typically require hundreds of episodes and minutes to hours of execution, stressing long-horizon planning, long-context management, and iterative debugging rather than one-shot problem solving. We evaluate 15 frontier models and find that agents consume on average 9.9M tokens per task, with roughly 231 episodes and 85.3 minutes of execution time per run, making Long-Horizon-Terminal-Bench more demanding than prior terminal-based benchmarks. Even the strongest tested model achieves 15.2% pass@1 at a partial-reward threshold of 0.95 and 10.9% at a perfect-reward threshold of 1.0, while the mean pass rate across models is 4.3% and 1.7% under the two thresholds, respectively. These results reveal headroom for improvement. We further analyze failure modes and error patterns, and release Long-Horizon-Terminal-Bench to support future progress on long-horizon terminal agents.

In long-horizon tasks, decision-relevant state is often scattered across an expanding trajectory, while the action agent must surface it and act. As trajectories grow, task requirements, environment facts, prior attempts, diagnoses, and open subgoals can be buried in the context window or pushed beyond it, failing to influence decisions when needed. We call this failure mode "behavioral state decay". We study memory as an active intervention mechanism rather than passive retrieval. A separate memory agent runs alongside an unmodified action agent, updating a structured memory bank from the recent trajectory and deciding whether to inject a memory-grounded reminder or remain silent. The module is plug-and-play with frontier action agents and existing agent harnesses. Across Terminal-Bench 2.0 and $τ^2$-Bench, it improves pass@1 for both weaker and stronger action agents, with gains of +8.3 pp on Terminal-Bench and +6.8 pp on $τ^2$-Bench. Ablations show that selective intervention outperforms passive bank exposure, always-on injection, advisor-only guidance, and general retrieval. As an early step toward open-weight memory policies, we train Qwen3.5-27B on SETA using SFT and GRPO, improving validation reward and achieving partial transfer to Terminal-Bench.

The optimization of long-horizon agents increasingly relies on reflection-based mechanisms, where a large language model (LLM) acts as an optimizer to diagnose agent failures and improve agent policies. However, real execution traces are difficult to use directly for optimization: large trace collections are often redundant and heterogeneous, making optimization inefficient and prone to overfitting to low-value failures; meanwhile, each individual trajectory also contains many irrelevant steps, while naive context reduction methods such as truncation or sliding windows can discard causally important evidence and produce misleading optimization signals. To resolve this dilemma, we introduce STRACE (Structural TRajectory Analysis and Causal Extraction), a framework that constructs high signal-noise optimization contexts for more precise and effective optimization. At the batch level, STRACE mines failure patterns to filter redundant traces and retain representative failures; within each selected trace, it performs causal localization over a textual dependency graph to remove non-causal steps and identify the true root-cause module for optimization. Empirical results demonstrate that STRACE significantly outperforms standard context-filtering baselines. Notably, on a challenging formal verification task (VeruSAGE-Bench), it successfully optimizes human-expert designed agents, delivering $1.4\times$ success-rate improvement (42.5% to 58.5%). The code is available at https://github.com/moomight/STRACE .

Reinforcement learning has become a promising paradigm for improving large language model (LLM) agents on long-horizon search tasks, where the agent must make a sequence of intermediate decisions before receiving a final outcome. However, existing methods still face a key limitation: the rollout budget is often allocated without explicitly assessing the utility of intermediate states. As a result, substantial computation may be spent on low-value states, even though different branches can vary drastically in their informativeness. In this paper, we propose Information Gain-based Rollout Policy Optimization (IGRPO), a policy optimization framework that treats intermediate-state informativeness as the organizing principle of rollout collection. Specifically, IGRPO performs budget-aware tree-structured rollouts by allocating expansion budget according to node-level informativeness, so that more informative branches are expanded more frequently while unpromising branches are progressively suppressed. We further demonstrate that the information gain-based rollout induces an explicit limiting teacher distribution over trajectories, which naturally yields a clear policy optimization target, thereby unifying adaptive tree-structured exploration with principled policy learning under a single framework. Experiments on seven challenging search-augmented QA benchmarks demonstrate that IGRPO consistently outperforms strong baselines under the same rollout budget constraints, validating the effectiveness of leveraging the induced teacher distribution to guide policy optimization for long-horizon search agents.

On-policy distillation (OPD) trains a student policy by matching a stronger teacher on the student's own trajectories, offering a promising framework for language agent training. However, its application to long-horizon agentic tasks remains insufficiently explored. We identify two key inefficiencies in vanilla agent OPD: (1) full-horizon rollouts often waste wall-clock resources on tail turns that provide weak and noisy KL supervision, and (2) trajectory-level KL objectives concentrate most of the loss on shallow tokens, leaving deeper decision turns under-trained once initial behaviors are aligned. To address these challenges, we propose TurnOPD, a turn-level budgeting strategy for efficient on-policy distillation of long-horizon agents. TurnOPD consists of two budget controllers: adaptive rollout-depth budgeting, which uses probe-based turn statistics to determine rollout length, and progressive turn-normalized loss budgeting, which gradually shifts KL weighting from token-level to turn-balanced supervision. Experiments on ALFWorld, WebShop, and Multi-Hop Search with task-specialized teacher models show that TurnOPD achieves superior validation accuracy under equal wall-clock training budgets and advances the accuracy--time frontier beyond vanilla OPD.

Long-horizon agentic LLMs are increasingly limited by finite context windows, as extended interaction trajectories can exceed the maximum context length before a task is completed. Context compaction offers a natural solution by summarizing previous interaction states and continuing the rollout under a compressed context, but incorporating compaction into reinforcement learning remains underexplored. We propose CompactionRL, a reinforcement learning strategy to train long-horizon agentic LLMs with context compaction. Our approach jointly optimizes task execution and summary generation with token-level loss normalization and cross-trajectory generalized advantage estimation. This design enables the LLM agents to learn from compacted long-horizon trajectories. We train CompactionRL on top of open models and observe consistent performance gains on agentic coding tasks. CompactionRL enables the open GLM-4.5-Air model (106B-A30B) to achieve Pass@1 scores of 66.8% on SWE-bench Verified and 24.5% on Terminal-Bench 2.0, with absolute gains of 7.0 and 3.1 points, respectively. Built upon GLM-4.7-Flash (30B-A3B), CompactionRL improves Pass@1 by 5.5 and 6.8 points, reaching 56.0% on SWE-bench Verified and 20.2% on Terminal-Bench 2.0, respectively. CompactionRL is thus deployed in the RL pipeline for training the open GLM-5.2 model (750B-A40B).

Reinforcement Learning (RL) is the dominant paradigm for training Large Language Model (LLM) agents on long-horizon tasks. However, sparse and delayed rewards often lead to trajectory neglect, in which agents lose focus on the task goal and interaction history at intermediate steps. Prior work has explored step-level supervision using Shannon-entropy-based uncertainty signals, which conflate inherent state complexity with agent confidence and therefore provide unreliable estimates of decision reliability. To address this issue, we propose normalized entropy, which measures confidence deviations relative to an agent's average behavior under a given state, thereby strengthening the association between low-quality actions and trajectory neglect. Building on this insight, we introduce Selective Trajectory-Aware Policy Optimization (STAPO), a hierarchical group-based RL framework. STAPO leverages normalized entropy to locate outlier steps associated with trajectory neglect and optimizes them via a joint mechanism of trajectory-aware reward and trajectory-independent penalty, enhancing trajectory awareness while preserving training stability. Extensive experiments on ALFWorld, WebShop, and Search-Augmented QA demonstrate that STAPO achieves state-of-the-art performance while substantially alleviating trajectory neglect, validating its effectiveness and robustness for agentic tasks.

Long-lived AI agents require continuity across interactions, but continuity cannot be obtained by simply extending the prompt window. An agent must preserve useful prior experience, retrieve it selectively, distinguish personal context from external evidence, and revise memory when the underlying situation changes. We propose an architectural memory substrate organized along two orthogonal axes: a representational axis spanning structured records, vector representations, and graph relations; and a temporal axis spanning short-term traces, medium-term abstractions, and long-term semantic commitments. Its key design constraint is synchronized structured-vector-graph memory: structured records govern eligibility, vector representations support recall, and graph relations adjudicate support, contradiction, and supersession before gated context projection. Its central claim is that reliable personalization is a memory design problem: useful memory is structured, selectively exposed, continuously consolidated, and epistemically labeled rather than stored as undifferentiated conversation history. Beyond the framework, we instantiate MRMS as a lightweight prototype implementing structured records, vector retrieval, temporal policies, and graph-based revision. The prototype exercises the core substrate mechanisms through pre-generation memory selection, revision, boundary enforcement, and evidence attribution under controlled long-lived interaction scenarios with explicit evidence requirements.

Lifelong agents need more than larger context windows and better retrieval. They need memories that can persist, evolve, and be corrected without forcing the serving stack to recompute the same history on every turn or silently reuse stale runtime state. We present PLACEMEM as a systems position on lifelong-agent memory, instantiated by an executable control-plane prototype. The central claim is that agent memory should be represented as versioned capsules that unify semantics, provenance, validity, and reusable runtime state under one correction-aware identity. In the current prototype, capsules drive prompt-level text retrieval, KV-aware routing, and cascading invalidation over live streamed backends; prospective layer-frontier replay is intentionally framed as a deeper integration agenda rather than a claimed engine feature. We describe a vLLM-first prototype with persistent capsule state, concurrency-safe invalidation, an OpenAI-compatible routing sidecar, a typed metadata contract, and a benchmark harness that measures live first-token latency, reuse, and post-correction behavior. The result is both an executable artifact that demonstrates correction-aware control-plane behavior today and a concrete roadmap for replay-aware serving integration in future lifelong-agent systems.

Long-horizon LLM agents accumulate tool results, files, plans, and user constraints that are too structured to be treated as a disposable text suffix. Current systems mostly rely on in-run heuristics such as chronological pruning and tool-output masking, or on final self-summary near a context limit. Heuristics are cheap but blind to future dependencies; summaries preserve narrative state but often hide exact evidence, locators, and editable artifacts. We present Self-GC, where GC denotes self-governing context while deliberately echoing garbage collection: the system does not merely reclaim unused tokens, but governs the lifecycle of agent context objects. Self-GC turns user turns, tool spans, and skill state into indexed objects; asks a side-channel planner to propose fold, mask, and prune actions; and lets the harness enforce recoverable sidecars, safe commit boundaries, and cache-aware commit. On a 33-session Hard Set, Self-GC prunes 43.95% of prefix tokens while leaving 84.85% of future continuations unaffected, compared with no-impact rates of 54.55% to 69.70% for heuristic baselines. On a 332-session production-derived suite, three planner backbones reach no-impact rates of 91.27% to 94.58%, while baselines remain at 77.71% to 87.46%. In production, an online account-level split reduces daytime average input tokens by 10% to 15%, with peak reductions near 20%. These results point to context management as runtime lifecycle control over indexed, recoverable objects rather than post hoc text cleanup.

While long-horizon mobile GUI agents typically rely on thought-action-observation loops, they struggle to separate persistent task states from transient screen observations. As execution histories grow, this entanglement imposes a severe context burden, causing agents to forget initial requirements, hallucinate progress, or repeatedly interact with stale interfaces. To address this, we introduce Task-State Representation (TSR), a training-free framework that explicitly decouples task state from sensory input. Acting as a lightweight external wrapper, TSR maintains three structured components: a global instruction summary, a dynamic progress tracker for subgoals, and a transition-aware action verifier. By continuously updating through pre- and post-action visual comparisons, TSR effectively guides the agent's reasoning without requiring architectural modifications. Experiments across four mobile GUI benchmarks validate TSR's effectiveness, yielding up to a 12 absolute point increase in success rate on complex cross-application and memory-intensive tasks.

Long-horizon language agents must repeatedly interact with tools, accumulate evidence, and make decisions under bounded context windows. Context-management methods make such rollouts feasible by simplifying past interactions through deletion, folding, or memory editing. However, when useful history is collapsed into compressed states, the reconstructed context may no longer reveal which earlier observations support a successful final answer. This creates a mismatch between bounded-context acting and outcome-based reinforcement learning: the policy acts on reconstructed context, while the learner lacks source-level provenance for assigning credit to the evidence that mattered. We propose ECHO, a selective turn-memory framework for traceable context reconstruction in Agentic RL. ECHO compresses each completed environment turn into a compact source-indexed memory record, reconstructs bounded policy contexts by selecting useful records, and reuses the selected source indices to route positive outcome credit to the final trajectory segment, reused evidence turns, memory findings, and memory-selection actions. On BrowseComp-Plus, ECHO reaches 43.4% held-out accuracy, outperforming GRPO at 28.9% and the rolling-summary baseline SUPO at 36.1%, while using fewer turns and lower trajectory volume than SUPO. The trained policy also improves zero-shot generalization across multi-objective QA, code generation, and deep information-seeking benchmarks on both dense and MoE backbones.

Mobile GUI agents increasingly face long-horizon tasks that require reading, updating, and reusing task-relevant data across pages and applications. Existing methods treat memory largely as passive storage, where past observations are accumulated and retrieved when needed. Yet retrieving a value does not reveal its current role in the workflow. The agent must still infer from accumulated records whether the value should be used now, has already been used, or must wait for a later dependency. This implicit reconstruction becomes unreliable in long trajectories with repeated values, distractors, and outdated states, causing repeated or missed operations. To address this, we propose Active Task Driving Memory (ATMem), which shifts GUI-agent memory from passive storage to an actively maintained execution state. ATMem maintains task-relevant information as a continually updated execution state that links each value to its role and current status, enabling action selection based on the current workflow state. While supervised fine-tuning enables the agent to construct ATMem, it does not teach when ATMem is beneficial. We therefore introduce STR-GRPO, an online reinforcement learning method that encourages selective use of ATMem based on its contribution to task completion. STR-GRPO contrasts memory-on and memory-off rollouts to estimate when memory use improves execution, while memory-cost-aware reward discourages costly memory usage that does not improve execution. To evaluate whether agents can complete all in-scope work while avoiding out-of-scope actions, we build a challenging mobile benchmark. From a list of near identical entries, agents must act on every entry that satisfies the instruction and reject entries that violate its constraints. We further introduce App-Level Progress and Scope-Aware F1 to measure these two dimensions separately.

Water looks unchanged as it warms, then at a critical point it boils. We ask whether long-horizon language agents show an analogous transition in their implicit world models. In some parameter settings, changing state load by a small amount, or adding a single step of horizon, leaves behavior nearly unchanged; near a critical boundary, the same small change causes a sudden world collapse. We study this effect in a deterministic task family with exact per-step gold state. A large grid search over state cardinality, dependency density, horizon, branching, observation mode, and mutation rate reveals a phase diagram: a solved plateau, a narrow transition band, and a collapse floor. Per-step traces show the mechanism: world-state fidelity fails before action validity, so the agent is not merely choosing a bad action; it is acting from a corrupted world. Stronger models translate the critical boundary but do not remove the qualitative transition. These results make world-model collapse a measurable bottleneck for long-horizon agents.

We introduce Agents-A1, a 35B Mixture-of-Experts Agentic Model that reaches trillion-parameter-level performance by scaling the agent horizon. We investigate agent-horizon scaling from two perspectives: scaling long-horizon trajectories and scaling heterogeneous agent abilities. To support this goal, we build a long-horizon knowledge-action infrastructure that connects external knowledge, actions, observations, and verifier outcomes, producing agentic trajectories with an average length of 45K tokens. Based on this, we train Agents-A1 with a three-stage recipe. First, we perform full-domain supervised fine-tuning to align the base model with broad agentic behaviors. Second, we train domain-level teacher models to capture specialized expertise in each domain. Third, we propose a multi-teacher domain-routed on-policy distillation with salient vocabulary alignment to improve knowledge transfer efficiency across different domains, unifying six heterogeneous domains into one deployable student model. Agents-A1 achieves strong and broad performance for long-horizon agent benchmarks. Compared with 1T-parameter model such as Kimi-K2.6 and DeepSeek-V4-pro, Agents-A1 achieves leading results on SEAL-0 (56.4), IFBench (80.6), HiPhO (46.4), FrontierScience-Olympiad (79.0), and MolBench-Bind (56.8), and remains highly competitive on SciCode (44.3), HLE (47.6) and BrowseComp (75.5). We hope this work provides the community with a practical path for scaling the horizon using a 35B agent that can reach or match the performance of 1T models on long-horizon tasks.

Understanding and navigating human-centered environments over extended periods of time while considering human behavior and routines remains a fundamental challenge in robotics. In real-world settings, robots may be asked to locate a specific individual, predict where that person is likely to be, or estimate when they typically leave a building. Addressing such queries requires reasoning over extensive histories of observations and capturing long-term behavioral patterns. To this end, we introduce Human-Centered Memory for Embodied Robots (HUMEMBR), a system designed for embodied question answering and routine-conditioned navigation. HUMEMBR integrates a continuous memory construction process with a parallel retrieval and querying mechanism, enabling the system to accumulate structured representations of human routines while supporting interactive, user-driven queries. Our experimental results indicate that HUMEMBR improves long-horizon reasoning about human behavior relative to full-context LLM baselines, while using substantially fewer tokens. Furthermore, we deploy HUMEMBR on a physical robot in two distinct environments, showing its ability to handle diverse queries and navigation tasks under real-world conditions.

Long-horizon strategic planning in complex strategy games requires coordinating tightly coupled decision domains, including technology, economy, diplomacy, and military, across hundreds of turns under imperfect information. Existing LLM-based agents face three challenges in this setting. They recover little relational structure, such as distance or threat, from raw coordinate observations. They serialize the entire growing game state into every prompt and plan all domains in a single output, so the context eventually overflows and urgent domains crowd out long-term ones. The only reward is a lagged final score, which provides no progress signal within or across games.<br/><br/>We present SAGA, an LLM multi-agent framework that addresses these challenges with three mechanisms: a Map-Semantic Scene Graph that renders typed entity relations as concise per-entity text, a Tool-Augmented Planner that retrieves state on demand and routes disjoint per-domain directives to specialist controllers, and a Dual-Horizon Feedback Loop that sets intermediate goals within a game and distills causal lessons across games. On CivRealm, a benchmark built on the strategy game FreeCiv, SAGA attains the highest mean final score among six methods, the statistically strongest gains on infrastructure, and a 27% reduction in output tokens. With cross-game evolution enabled, it reaches the highest score over five successive games.

The database community has repeatedly advanced the state of the art by recognizing that new workloads demand new system architectures. We argue that long-horizon agentic tasks -- code generation, scientific discovery, hardware design -- are such a workload. These agents explore: they generate artifacts, execute tools, observe failures, branch, and repair over hundreds of steps. This search produces a structured object we call an experience graph: executable artifacts, tool outputs, rewards, sibling comparisons, and causal lineage. Yet existing agent frameworks treat this experience as disposable state -- JSON checkpoints and session logs that cannot be recovered after a crash, queried across users, or materialized into training data. We propose Trellis: a data foundation that treats the experience graph as first-class, governed, queryable database state. The core insight is that search over experience graphs is a database access pattern. Frontier selection is a query, cross-session reuse is vector-seeded graph retrieval, training-data extraction is a materialized view, and reconstructing what an agent knew at any past step is a time-travel query. When the database owns the experience graph, agents become stateless compute, and crash recovery, horizontal scaling, and a closed-loop training flywheel emerge as architectural byproducts. We ground the design in KernelEvolve, a production accelerator-kernel optimizer at Meta, where cross-session reuse reaches a target speedup roughly 10x faster at 52% lower token cost. More broadly, Trellis turns inference-time search from disposable computation into a durable institutional asset: logs made databases reliable; experience graphs may make agents cumulative.

Long-horizon software engineering agents often need to manage lengthy and noisy interaction histories under limited context budgets. Existing memory management methods typically rely on static compression workflows or impose rigid constraints on compression timing and granularity. Moreover, these approaches fail to jointly optimize memory management and issue resolution capabilities to improve performance while reducing token usage. We present SWE-MeM, a training framework for proactive and on-demand memory management in software engineering agents. SWE-MeM provides a flexible memory tool that lets agents decide when, what, and how to compress based on trajectory state, task progress, and remaining context budget. We train agents with synthesized proactive memory-management trajectories and Memory-aware GRPO, which jointly optimizes memory management and issue resolution through memory-aware trajectory splitting and step-level credit assignment. On SWE-Bench Verified, SWE-MeM achieves 43.4% and 60.2% resolve rate with 4B and 30B models, respectively, outperforming existing memory management baselines in both performance and efficiency.

Large language models can serve as capable long-horizon agents, but their out-of-distribution (OOD) generalization remains weak. We identify a key source of this failure as task insensitivity: when faced with similar but distinct tasks, models might apply patterns learned during training and fail to solve the task at hand. We show that models often continue with actions aligned with the original task even when the instruction is semantically corrupted and cannot be directly answered. We further find that, when we replace the task description in a trained prompt with another similar but distinct task, the model may still output the same action. This behavior is accompanied by a consistent training-time attention drift away from task tokens and toward local observations, suggesting an optimization bias toward shortcuts. To mitigate this problem, we propose Task-Perturbed NLL Optimization, a lightweight contrastive regularizer that explicitly encourages action dependence on the task instruction. Extensive evaluations show that our intervention improves task sensitivity and OOD generalization while preserving more stable attention to task tokens.

Long-running language agents need more than memory access. Retrieval systems can fetch past facts at query time, but they do not decide which experiences should continue to shape behavior after the working context is unloaded. We study this separate problem as memory depth: durable goal-conditioned tendencies written into a small parametric store. We introduce the loop-drift protocol, a controlled stress test in which the retrieval index remains intact while working context is unloaded and goal-conditioned behavior must persist under long-loop interference. We evaluate EVAF, a surprise- and valence-gated LoRA consolidation mechanism. Across GPT-2 and TinyLlama, retrieval is strongest on shallow factual recall (short-fact accuracy 0.956--0.973), while EVAF is strongest on goal persistence and post-unload recovery (0.812--0.904) with only 2--3 parametric writes per 200 events. Mechanism controls show that selective consolidation factorizes into two controllable dimensions: selection and actuation. Matched random gates isolate selection beyond sparse writing; fixed-inner controls across GPT-2, TinyLlama, and Mistral-7B show that inner-loop write strength is model-dependent; and a Mistral-7B matched-gate inversion reveals asymmetric selection-actuation coupling under miscalibrated actuation. Public Memora event streams serve as an external diagnostic, exposing stale-memory invalidation as an unresolved boundary. Within this probe, selective parametric consolidation supplies memory depth distinct from and complementary to retrieval access.

As autonomous agents tackle increasingly complex multi-step, multi-agent tasks, their execution trajectories have scaled beyond the constraints of even the largest context windows. Current methods for effectively diagnosing agent failures load the full trajectory into an LLM's context window, which suffers from attention dilution and fails when agentic traces inevitably exceed context limits. To address this, we introduce SAFARI (Scaling long-horizon Agentic Fault AttRibution via active Investigation), a framework that replaces linear context loading with a tool-augmented diagnostic loop. By equipping LLMs with a specialized toolbox to read and search trajectory segments alongside a persistent Short-Term Memory (STM) for cross-turn reasoning, SAFARI effectively decouples diagnostic accuracy from architectural context limits. Our experiments demonstrate that SAFARI outperforms state-of-the-art results by 20% on the Who&When dataset within a 1M token budget, and by 19% on TRAIL GAIA subset on a 25K token budget. Most significantly, SAFARI maintains a 0.58 precision even when the target fault resides 5x beyond the model's native context window, a scenario where traditional evaluators fail entirely.

As large language model (LLM) agents are applied to longer tasks, they increasingly modify workspace state across multiple rounds of iteration. However, agents typically observe only tool outputs and log fragments, while the actual state changes occur in the file system. Without explicit workspace boundaries, state-changing operations such as file writes and temporary artifact generation may scatter changes across paths. Over time, these weakly constrained changes accumulate, making states such as modified files difficult to track. This paper presents LemonHarness, an integrated execution framework for long-horizon agents. LemonHarness establishes an explicit execution boundary by constraining state-changing operations within a clearly defined workspace and bringing model invocation, tool execution, and rule knowledge within a single controlled boundary. State-changing operations, including file writes, dependency installation, and temporary artifact creation, are executed through structured tool interfaces, with execution feedback recorded as observations available to subsequent model decisions. The system also introduces a reusable rule knowledge base, which turns recurring execution rules and acceptance criteria into runtime knowledge. LemonHarness further adds a time-aware execution mechanism that exposes elapsed and remaining budget to the model, so it can rebalance exploration, implementation, and validation effort as time pressure shifts and avoid timeouts from long waits or excessive verification. On Terminal-Bench 2.0, LemonHarness_GPT-5.3-CodeX reached 84.49% accuracy over 445 trials; pairing the same framework with the stronger GPT-5.5 backbone raised the average accuracy to 86.52% across five jobs. The results suggest that a unified runtime boundary, callable rule knowledge, and time-aware execution can improve the stability of long-horizon agent execution.

Modern LLM agents increasingly rely on context compaction, summarization, or eviction to keep long-running sessions within a token budget. We show that this context-management layer is a safety-critical failure surface: in-context governance constraints that agents reliably obey while visible can be silently removed by compaction, causing the same agent to perform prohibited tool actions later in the session. We call this failure mode Governance Decay. We introduce ConstraintRot, a benchmark of long-horizon agent scenarios with deterministic tool-call grading, and measure compaction-induced violations across seven model families. Across 1,323 episodes, violation rises from 0% with the policy in full context to 30% after compaction, reaching 59% for some models; when the constraint survives the summary, violation remains 0%, but when it is dropped, violation reaches 38%. We further study a Compaction-Eviction Attack, in which adversarial in-context content biases the summarizer to omit a legitimate policy, and show that optimized injections defeat every evaluated model. Finally, we propose Constraint Pinning, a simple training-free mitigation that quarantines governance constraints from lossy compaction and restores violation to 0% in our benchmark. These results identify context management as a first-class governance surface for deployed LLM agents.

Long-running language-model systems accumulate interaction history that outgrows the context window, so they must continually evict. When an eviction policy drops a task-critical detail, for example an access token issued at login or a path the next call needs, the action fails. We present LRE (Learned Relevance Eviction), a kilobyte-scale, CPU-only, language-model-free scorer that learns which units of history are task-critical and keeps them by verbatim extraction. Under a matched-budget comparison, in our experiment, no baseline dominates LRE on the accuracy-cost plane. On agents, LRE recovers 93% of the aggregate accuracy of keeping the entire history (41.1 vs. 44.0) and exceeds it by 27% on the simplest tasks, while requiring zero compressor calls and cutting the worst-case peak prompt by 52%. A controlled study trace shows LRE completes tasks where the others loop, finishing one such task in 37% fewer calls than keeping everything and solving 14 tasks where no other run policy does. On conversational memory, LRE outranks dense and token-pruning encoders at zero neural cost while being 295-1569x smaller in size. In downstream evaluation, LRE gives the best budgeted answer quality on LoCoMo reading 68% fewer tokens. Its supervision can also be annotation-free: training only on the system's own behavior recovers 95% of the supervised scorer's effectiveness. We argue that, because memory eviction in LLM agents is a fidelity problem, it requires a deployable proactive policy where the future query is unavailable and exact state is decisive, and that cheap learned relevance can be sufficient.

To assist humans over extended periods in real homes, embodied agents must remember user routines, world states, and past interactions. Existing long-term memory benchmarks mainly evaluate language-centric retrieval and question answering, while embodied benchmarks often focus on short-horizon task execution without testing long-term memory use in dynamic environments. We introduce WorldLines, a project-driven benchmark for long-horizon embodied household assistance. It constructs temporally extended household traces with dialogues, actions, execution feedback, object and device state changes, and converts them into evidence-linked samples for Memory QA and Embodied Task Planning. We further propose ObsMem, an observer-grounded memory framework that maintains visibility-aware memories and action-native state trails for state-aware decisions. Experiments reveal persistent challenges in partial observability, overwritten world states, and translating long-term memory into embodied plans, while ObsMem offers a stronger reference architecture for this setting.

Language model agents are becoming proficient executors at isolated, short-horizon tasks such as software engineering and customer service. Yet real-world challenges require a combination of sophisticated skills that remain largely untested in agents: (1) navigating long horizons amid uncertainty; (2) acquiring information in noisy environments; (3) adapting to a changing world; (4) orchestrating multiple moving parts toward a coherent goal. We introduce CEO-Bench, which evaluates these capabilities together by simulating a representative real-world task: operating a startup for 500 days. An agent manages pricing, marketing, budgeting, and many other aspects of a fictional company through a programmable Python interface, operating in the same environment and facing the same challenges as a human CEO. Success demands analyzing noisy, interconnected business databases, translating signals into sound strategy, and coordinating many decisions with programming. The strongest agents write sophisticated code that simulates customer cohorts to forecast future cash and mines negotiation history to uncover hidden customer preferences. Even so, most state-of-the-art models struggle in this environment. Only Claude Opus 4.8 and GPT-5.5 finish above the $1M starting balance, and neither consistently turns a profit. CEO-Bench takes a first step toward measuring the intelligence required to drive sustained, adaptive progress over time.

Reinforcement learning (RL) has emerged as a powerful paradigm for training Large Language Models (LLMs) as agents. However, conventional RL methods for long-horizon agentic tasks often struggle with sparse outcome rewards. Intuitively, this overlooks the rich environment dynamics information contained in rollout interaction trajectories. We argue that the interaction experience inherently serves as an implicit supervision signal, reveals the underlying transition mechanisms of the environment, and enables the agent to construct a more accurate internal model of the environment.. Therefore, in this work, we investigate how to leverage this additional signal to improve policy learning. Specifically, we propose EnvRL, a framework that incorporates environment dynamics learning into agentic RL via two auxiliary objectives: state prediction and inverse dynamics. By jointly optimizing with the primary RL objective, we encourage the agent to internalize environment dynamics from its own interaction experience. Extensive experiments on two long-horizon agentic benchmarks demonstrate that EnvRL achieves significant improvements on success-rates over RL-only baselines, e.g., when trained with GRPO, lifting Qwen-2.5-1.5B-Instruct from 72.8% to 77.4% on ALFWorld, and from 56.8% to 67.0% on WebShop.

Long-horizon agents rely on memory mechanisms to compress interaction history, but optimizing memory writing faces a distinct credit assignment challenge: a memory update may be rewarded or penalized due to downstream tool failures, noisy observations, or reasoning errors rather than its own contribution. We propose HiMPO, a Hindsight-Informed Memory Policy Optimization framework for assigning less-entangled credit to memory-writing actions in long-horizon agents. HiMPO first estimates the local utility of a memory update by comparing the task-relevant information recoverable from the previous and updated memories under the same pre-write state. It then uses hindsight relevance as a bounded retrospective filter that attenuates memory credit when local utility is not supported by the target outcome. The resulting memory-specific advantage is applied only to memory tokens, while trajectory-level rewards optimize the rest of the agent's behavior. Across judge-based open-domain tasks and objective compressive-memory QA, HiMPO improves over strong memory-based and RL-based baselines while preserving compressed-context efficiency. Controlled interventions and live replay studies further show that HiMPO reduces blame leakage from tool-induced errors, assigns memory credit that aligns with the functional impact of memory writes, and remains robust to noisy training targets.

Deep research agents aim to solve complex knowledge-intensive tasks through long-horizon planning, evidence gathering, reasoning, and report generation. While recent progress in search agents has demonstrated strong capabilities in information retrieval and answer verification, most existing training datasets remain search-centric, focusing primarily on closed-ended question answering and information localization. As a result, they mainly train information-seeking behavior while providing limited coverage of key deep research capabilities, including evidence integration, knowledge synthesis, planning, file understanding, and structured report generation. In this work, we propose a unified trajectory construction paradigm for deep research agents that combines closed-ended QA and open-ended exploration. The proposed framework consists of graph-grounded task formulation, agentic trajectory rollout, and multi-dimensional trajectory verification, enabling scalable synthesis of high-quality agentic trajectories spanning long-chain complex reasoning, deep research instruction following, report writing, file understanding and generation, and skills usage. Compared with existing search-oriented datasets, our synthesized trajectories place greater emphasis on knowledge synthesis, complex reasoning, and planning. S1-DeepResearch-32B achieves state-of-the-art performance among open-source models of comparable scale across 20 benchmarks spanning five capability dimensions, including complex reasoning, instruction following, report generation, file understanding, and skills usage. On several challenging deep research benchmarks, it approaches the performance of leading proprietary frontier models. These results highlight the importance of jointly modeling information acquisition, knowledge synthesis, and planning-oriented agent behaviors for building effective deep research agents.

A central role of personal-agent memory is to turn stored information and prior interactions into future-oriented assistance. In daily use, useful cues come from what the agent observes and how the user interacts with the agent, and the agent must carry them forward from the current request to similar future tasks. Existing memory benchmarks usually test dialogue recall or task improvement in isolation, leaving the trajectory from streaming observations to later assistance largely untested. We introduce StreamMemBench, a streaming benchmark that constructs a two-step task sequence around each evidence anchor from EgoLife egocentric streams. The initial task tests evidence use, while the follow-up task tests whether feedback and interaction experience are reused. Four metrics diagnose evidence recall, initial evidence use, feedback incorporation, and follow-up reuse. Experiments with eight memory systems across two backbones show that current systems often fail to use observed evidence or turn feedback into reliable follow-up behavior, even when evidence is stored or feedback is incorporated locally. StreamMemBench is publicly available at https://github.com/landian60/StreamMemBench.

This paper presents a structured analysis of security challenges in long-horizon agentic AI systems. The study reviews existing threats, evaluation approaches, attack propagation mechanisms, and security frameworks. A taxonomy of security threats and a framework for analyzing attack propagation are proposed to support future research in agentic AI security

Rapid advances have been made in developing general-purpose embodied agent in environments like Minecraft through the adoption of LLM-augmented hierarchical approaches. Despite their promise, low-level controllers often become performance bottlenecks due to repeated execution failures. We argue that a key limitation is not only the lack of episodic memory, but also the decoupling of \textit{what-where-when} memory from \textit{which-why} reasoning. To address this, we propose \textbf{WISE} (Which-Why Informed Semantic Explorer), a long-horizon agent framework with an enhanced low-level controller equipped with a Causal Event Graph that augments episodic memory with explicit causal structure linking observations to task relevance. Unlike prior work such as MrSteve, which relies on feature similarity for retrieval, WISE enables robust recall under viewpoint changes and supports opportunistic task reordering through causal reasoning. Building on this memory, we propose an Opportunistic Task Scheduler that dynamically re-prioritizes subtasks when causally relevant opportunities are detected. We further equip WISE with a multi-scale progressive exploration strategy to provide spatially comprehensive observations for downstream reasoning. Experiments show that WISE largely improves task success and efficiency on long-horizon sparse tasks, particularly in settings requiring adaptive decision-making.

Search agent benchmarks exemplified by BrowseComp have rapidly saturated over the past year, with the strongest models surpassing 90% accuracy. Since these benchmarks are predominantly human-authored, annotators lack a global perspective on entity statistics and cannot systematically maximize search space size and structural complexity. This creates a difficulty ceiling that is hard to break. To address this, we introduce LoHoSearch (Long-Horizon Search Agents), a challenging benchmark comprising 544 human-verified questions across 11 domains. LoHoSearch is constructed via an automated pipeline built upon a knowledge graph covering over 7 million Wikipedia entities, which selects relations with large search spaces and assembles them into structurally complex questions with KG-verified unique answers. Our evaluation demonstrates that even the strongest model achieves only 34.74% accuracy, and existing context management strategies (best +6.8%) yield far smaller gains than on prior benchmarks. LoHoSearch provides a more demanding standard for evaluating long-horizon reasoning and context management in search agents.

Long-horizon tool-use reinforcement learning learns from outcome verification, but trajectory-level advantages are broadcast over reasoning, API, and answer tokens. Direct self-distillation can supply a denser signal, but in our experiments it can also destroy tool use by rehearsing teacher behavior without identifying which actions the verifier rewards. We introduce Sibling-Guided Credit Distillation (SGCD), which uses distillation for bounded credit weighting rather than as a competing actor loss. Dynamic sampling produces mixed successful and failed sibling rollouts; an external LLM summarizes their contrast into a training-only credit reference; and detached teacher/student divergence reshapes GRPO token advantages. The deployed student receives only the clean task prompt. Across AppWorld and tau^3-airline, SGCD reports higher held-out point estimates than GRPO-family comparators: AppWorld TGC improves from 42.9 to 45.6 on test_normal and from 24.7 to 27.0 on test_challenge, and tau^3-airline held-out evaluator score improves from 0.583 to 0.602. These results support a narrow design rule for long-horizon tool-use agents: use distillation to guide credit assignment while keeping policy gradient in charge of the actor update.

Large language model (LLM) agents struggle with long-horizon tasks due to their inherent statelessness, requiring all task-relevant information to be encoded in growing input contexts. The resulting degraded reasoning quality, increased inference cost, and higher latency necessitate efficient working memory mechanisms. However, existing approaches either rely on lossy compression or similarity-based retrieval, which often fail to capture temporal structure and causal dependencies required for multi-step agentic tasks. In this work, we present HORMA, a Hierarchical Organize-and-Retrieve Memory Agent that organizes experience into a file-system-like hierarchical structure, where summarized entities are linked to the corresponding raw trajectories, enabling efficient access without losing detailed information. HORMA decomposes working memory into two stages: structured memory construction and navigation-based retrieval. The construction module iteratively refines how experiences are structured by distinguishing between failures caused by missing information and those caused by misleading or overloaded context. The navigation module retrieves task-relevant context by traversing the hierarchy using a lightweight agent trained with reinforcement learning to select minimal yet sufficient context, thereby reducing latency along the critical execution path. Across ALFWorld, LoCoMo, and LongMemEval, HORMA improves task performance under constrained context budgets while requiring at most 22.17% of the baseline token usage in long conversation tasks. Compared to existing methods, it consistently achieves better efficiency-performance trade-offs and generalizes effectively to unseen tasks.

Long-horizon language agents accumulate observations, reasoning traces, and retrieved facts exceeding context windows, making memory retention a fundamental resource-allocation problem. Existing systems treat retention as local and do not model long-term consequences under observability constraints. To fill this gap, we formulate memory retention as a constrained stochastic optimization with budget feasibility, evidence utility, and delayed costs including miss, reacquisition, and stale penalties. We show this multi-step problem is NP-hard, making exact solution intractable. Moreover, deployment decisions must be made under partial observability. To address these challenges, we propose OSL-MR (Observability-Safe Learning for Memory Retention), a learning-augmented framework that enforces a strict separation between online-observable features and offline-available supervision. OSL-MR combines an evidence learner trained from realized evidence with a Mixed-Score heuristic that serves as a deployable online-safe baseline and an inductive prior. The policy learns query-conditioned evidence from interaction data and remains deployable under the same constraints. Experiments on LoCoMo and LongMemEval show OSL-MR outperforms recency-based, Generative Agents-style, and other heuristic baselines, especially under tight budgets. The Mixed-Score prior improves precision and recall, and sensitivity analysis shows robustness across cost settings. On small solvable instances, single-step optimization is insufficient to anticipate future demand shifts, while OSL-MR stays significantly closer to the dynamic-programming optimum, confirming the necessity of the sequential formulation and reinforcing our learning-guided approximation. These results establish constrained stochastic optimization and optimization-guided learning as a principled foundation for memory management in long-horizon agents.

Memory is essential for enabling large language model (LLM) agents to handle long-horizon reasoning tasks. Existing memory mechanisms are largely centralized, typically organizing retrieved information and interaction history within a single model context. This design imposes a fundamental trade-off: scaling reasoning trajectories risks context overload, whereas aggressive content pruning may result in irreversible information loss. Seeking a better trade-off, we draw inspiration from human cognitive systems, especially the functional complementarity between the prefrontal cortex (executive control) and the hippocampus (memory management), suggesting that such a trade-off need not be inherent, but may instead stem from centralized memory organization. To this end, we propose ActiveMem, a heterogeneous framework that decouples agent memory from the core reasoning process. Specifically, a high-level Planner utilizes distilled semantic gists to execute reasoning, while a lightweight, distributed memory system operates in parallel to actively accumulate and consolidate these gists throughout the task. Experiments on BrowseComp-Plus and GAIA show that ActiveMem achieves state-of-the-art accuracy with significantly reduced overhead, demonstrating the effectiveness of distributed active memory for long-horizon reasoning.

While Large Language Models (LLMs) have demonstrated strong capabilities as autonomous agents across a wide range of tasks, their performance often degrades in multi-turn long-horizon agentic tasks. Existing methods have made progress through fine-grained credit assignment to alleviate long-horizon sparse rewards and hierarchical reinforcement learning to decompose tasks and reduce long-term dependency. However, these methods still do not directly address long-context interference, in which continuously growing histories weaken the agent's ability to track the global task state and impair subsequent reasoning and decision-making. Inspired by the way humans handle complex tasks through subgoal decomposition and completed progress summarization, we propose Hierarchical Planning and Information Folding (HIPIF) for long-horizon LLM agent learning. HIPIF trains the agent end-to-end to organize long-horizon execution around explicit subgoals while folding completed subgoal histories to reduce long-context interference. Furthermore, to stabilize subgoal-based planning and execution, HIPIF combines hierarchical reflection and subgoal-oriented process rewards to guide subgoal generation, transition, and execution, without relying on costly auxiliary models or task-specific expert trajectories. Extensive experiments on three publicly available agentic benchmarks demonstrate the validity of our method.

Large language models are increasingly expected to handle complex, long-horizon real-world tasks whose context demands can grow without bound, yet model context windows remain inherently finite. Recent work explores a paradigm where a main agent decomposes tasks and dispatches subtasks to subagents, which execute and return only summarized results, conserving the main agent's context budget. However, performing this well requires delegation intelligence: the ability to decompose complex tasks, determine when and what to delegate, and integrate returned results into the ongoing workflow. Training data for this capability is scarce in naturally occurring text, and to our knowledge, how to synthesize such data and train models to acquire this capability remains largely unexplored in the open-source community. To bridge this gap, we present a preliminary exploration targeting deep research, a representative long-horizon agent task. Specifically, we design a harness that guides the model toward high-quality task decomposition and delegation, while constraining subagents to return results properly to support the main agent's workflow. The harness-guided trajectories naturally encode correct delegation decisions, which we use as supervised fine-tuning data to internalize delegation intelligence into model weights. Our resulting model, SearchSwarm-30B-A3B, achieves 68.1 on BrowseComp and 73.3 on BrowseComp-ZH, the best results among all models of comparable scale. We will release our harness, model weights, and training data to facilitate future research.