LLM-powered multi-agent systems can now automate the full research pipeline from ideation to paper writing, but a fundamental question remains: automation for whom? Researchers operate under different resource configurations, hold different methodological preferences, and target different output formats. A system that produces uniform outputs regardless of these differences will systematically under-serve every individual user, making personalization a precondition for research automation to be genuinely usable. However, achieving it requires three capabilities that current systems lack: accumulating reusable procedural knowledge across projects, retaining user-specific experience across sessions, and internalizing implicit preferences that resist explicit formalization. We propose NanoResearch, a multi-agent framework that addresses these gaps through tri-level co-evolution. A skill bank distills recurring operations into compact procedural rules reusable across projects. A memory module maintains user- and project-specific experience that grounds planning decisions in each user's research history. A label-free policy learning converts free-form feedback into persistent parameter updates of the planner, reshaping subsequent coordination. These three layers co-evolve: reliable skills produce richer memory, richer memory informs better planning, and preference internalization continuously realigns the loop to each user. Extensive experiments demonstrate that NanoResearch delivers substantial gains over state-of-the-art AI research systems, and progressively refines itself to produce better research at lower cost over successive cycles.
Deep Research agents driven by LLMs have automated the scholarly discovery pipeline, from planning and query formulation to iterative web exploration. Yet they remain constrained by a static, ``one-size-fits-all'' retrieval paradigm. Current systems fail to adaptively adjust the depth and breadth of exploration based on the user's existing expertise or latent interests, frequently resulting in reports that are either redundant for experts or overly dense for novices. To address this, we introduce Personalized Deep Research (PDR), a framework that integrates dynamic user context into the core retrieval-reasoning loop. Rather than treating personalization as a post-hoc formatting step, PDR unifies user profile modeling with iterative query development, dual-stage (private/public) retrieval, and context-aware synthesis. This allows the system to autonomously align research sub-goals with user intent and optimize the stopping criteria for evidence collection. To facilitate benchmarking, we release the PDR Dataset, covering four realistic user tasks, and propose a hybrid evaluation framework combining lexical metrics with LLM-based judgments to assess factual accuracy and personalization alignment. Experimental results against commercial baselines demonstrate that PDR significantly improves retrieval utility and report relevance, effectively bridging the gap between generic information retrieval and personalized knowledge acquisition. The resource is available to the public at https://github.com/Applied-Machine-Learning-Lab/SIGIR2026_PDR.
Vision-Language Models (VLMs) have demonstrated exceptional general reasoning capabilities. However, their performance in embodied navigation remains hindered by a scarcity of aligned open-world vision and robot control data. Despite simulators providing a cost-effective alternative for data collection, the inherent reliance on photorealistic simulations often limits the transferability of learned policies. To this end, we propose \textit{\textbf{S}andbox-\textbf{A}bstracted \textbf{G}rounded \textbf{E}xperience} (\textbf{\textit{SAGE}}), a framework that enables agents to learn within a physics-grounded semantic abstraction rather than a photorealistic simulation, mimicking the human capacity for mental simulation where plans are rehearsed in simplified physics abstractions before execution. \textit{SAGE} system operates via three synergistic phases: (1) \textit{Genesis}: constructing diverse, physics-constrained semantic environments to bootstrap experience; (2) \textit{Evolution}: distilling experiences through Reinforcement Learning (RL), utilizing a novel asymmetric adaptive clipping mechanism to stabilize updates; (3) \textit{Navigation}: bridging the abstract policy to open-world control. We demonstrate that \textit{SAGE} significantly improves planner-assisted embodied navigation, achieving a 53.21\% LLM-Match Success Rate on A-EQA (+9.7\% over baseline), while showing encouraging transfer to physical indoor robot deployment.
Ask ChatGPT about vacation planning, and it may infer your income. Ask it about medication, and it may infer your medical history. Because such inferences can expose more information than users intend to reveal, prior work argues that they are a defining privacy risk of LLM-based systems. Yet prior work has mostly shown that LLMs can make potentially violating inferences, not how users experience those inferences nor what controls users may want governing their use. We built the Reflective Layer, a visualization tool that surfaces example unstated inferences from users' own ChatGPT histories, and used it in a mixed-methods study with 18 regular ChatGPT users evaluating 215 surfaced inferences from their own conversations. Counterintuitively, participants reacted more strongly with curiosity and interest rather than distress and concern. Discomfort arose mainly when inferences felt misrepresentative of the user or misaligned with expected use. Participants were also markedly less comfortable with advertisers and third-party applications using those inferences than with platform providers. These findings suggest that the acceptability of LLM inferences is governed not only by its content, but by context-sensitive norms around how they are generated, retained within the platform, and transmitted beyond it.
Diffusion large language models (dLLMs) offer a promising paradigm for parallel text generation, but in practice they face an accuracy-parallelism trade-off, where increasing tokens per forward (TPF) often degrades generation quality. Existing acceleration methods often gain speed at the cost of accuracy. To address this limitation, we propose TAD, a Temporal-Aware trajectory self-Distillation framework. During data construction, we condition a teacher model on both the prompt and the ground-truth response to generate decoding trajectories, recording the intermediate masked states throughout the process. Based on how many decoding steps remain before each masked token is revealed, we partition masked positions into near and distant subsets. For near tokens, we train the student with a hard cross-entropy loss using the teacher trajectory tokens as labels, encouraging confident predictions for tokens that are about to be decoded. For distant tokens, we apply a soft KL divergence loss between the teacher and student token distributions, providing softer supervision and preserving future planning knowledge. This temporal-aware partition naturally gives rise to two deployment configurations: a Quality model that prioritizes accuracy and a Speed model that favors more aggressive acceleration. Experiments show that TAD consistently improves the accuracy-parallelism trade-off. On LLaDA, it raises average accuracy from 46.2\% to 51.6\% with the Quality model and average AUP from 46.2 to 257.1 with the Speed model. Our code is available at: https://github.com/BHmingyang/TAD
Epilepsy diagnosis and treatment require evidence-intensive reasoning across heterogeneous clinical knowledge, including biosignal patterns, genetic mechanisms, pharmacogenomics, treatment strategies, and patient outcomes. In this work, we present \textsc{EpiGraph}, a large-scale epilepsy knowledge graph and benchmark for evaluating knowledge-augmented clinical reasoning. \textsc{EpiGraph} integrates 48,166 peer-reviewed papers and seven clinical resources into a heterogeneous graph containing 24,324 entities and 32,009 evidence-grounded triplets across five clinical layers. Built upon this graph, \textsc{EpiBench} defines five clinically motivated tasks spanning clinical decision-making, EEG report generation, pharmacogenomic precision medicine, treatment recommendation, and deep research planning. We evaluate six LLMs under both standard and Graph-RAG settings. Results show that integrating \textsc{EpiGraph} consistently improves performance across all tasks, with the largest gains observed in pharmacogenomic reasoning (+30--41\%). Our findings demonstrate that structured epilepsy knowledge substantially enhances evidence-grounded clinical reasoning and provides a practical benchmark framework for evaluating knowledge-augmented LLMs in real-world neurological settings. Our code is available at: https://github.com/LabRAI/EEG-KG.
While Large Language Models (LLMs) have catalyzed progress in embodied intelligence, a fundamental gap between their inherent probabilistic uncertainty and the strict determinism and verifiable safety required in the physical world. To mitigate this gap, this paper introduces NEXUS, a modular framework designed for continual learning in embodied agents. Different from prior works that treat symbolic artifacts merely as static interfaces, NEXUS leverages them for symbolic grounding and knowledge evolution. The framework explicitly decouples physical feasibility from safety specifications: capability of agents is improved through closed-loop execution feedback, while probabilistic risk assessments are grounded into deterministic hard constraints to establish a rigorous pre-action defense. Experiments on SafeAgentBench demonstrate that NEXUS achieves superior task success rates while effectively refusing unsafe instructions, exhibiting robust defense against adversarial attacks, and progressively improving planning efficiency through knowledge accumulation.
Agentic large language models often rely on skills, reusable natural language procedures that guide planning, action, and tool use. In practice, skills are typically improved through prompt engineering or by aligning the task LLM itself, which is costly, model-specific, and often infeasible for closed-source models. Skill optimization is not a one-step problem but a recurrent process with two coupled levels of credit assignment: a useful skill must improve rollout quality under current conditioning, while a useful revision must turn observed outcomes into a better skill for the next round. We propose Skill-R1, a reinforcement learning framework for instance-level recurrent skill optimization from verifiable rewards. Rather than updating the task LLM, Skill-R1 trains a lightweight skill generator that conditions on the task context, prior rollouts, and their verified outcomes to produce skills that steer a frozen task LLM. This preserves black-box compatibility with both open- and closed-source models while making adaptation substantially cheaper than model-level updates. Skill-R1 proceeds over multiple generations: at each step, the current skill induces rollouts whose verified outcomes are fed back to produce the next revision. To optimize this recurrent process, we introduce a bi-level group-relative policy optimization objective combining intra-generation and inter-generation advantages. The intra-generation term compares rollouts under shared skill conditioning, while the inter-generation term rewards revisions that improve behavior across successive generations. Together, these provide a principled objective for directional skill evolution rather than one-shot self-refinement. Empirically, Skill-R1 achieves consistent gains over no-skill baselines and standard GRPO across benchmarks with verifiable rewards, with particularly strong improvements on complex, multi-step tasks.
Accurate short-term mobile traffic prediction is important for proactive resource allocation and low-latency network management in fifth generation (5G) and sixth generation (6G). While large language models (LLMs) can perform in-context learning (ICL) without task-specific retraining, naive ICL prompting may suffer from numerical instability and limited temporal reasoning when traffic dynamics fluctuate rapidly. In this paper, we propose a chain-of-thought (CoT)-enabled LLM-based mobile traffic prediction framework that operates in two phases: (i) an offline phase that constructs structured CoT demonstrations by generating rationales via a plan-based CoT (PCoT) pipeline (lecture, plan, and rationale), and (ii) an online phase that performs close to real-time prediction by retrieving the most relevant demonstrations using a similarity policy that considers both the historical throughput pattern and its short-term changes. We evaluate the proposed framework using a real-world 5G measurement dataset that includes both driving and static scenarios across diverse applications. Our numerical results reveal that the proposed 2-shot CoT-LLM can improve mean absolute error (MAE), root mean square error (RMSE) and R2-score by up to 14.88%, 15.03%, and 22.41%, respectively, compared to the 2-shot ICL-LLM and classical baselines. Furthermore, by optimizing the number of in-context examples, we achieve additional improvements of 4.58%, 5.70%, and 4.85% in MAE, RMSE, and R2-score, respectively.
The Model Context Protocol (MCP) has unified the interface between Large Language Models (LLMs) and external tools, yet a fundamental gap remains in how agents conceptualize the environments within which they operate. Current paradigms are bifurcated: Task-level planning often ignores execution-time dynamics, while reactive execution lacks long-horizon foresight. We present MCP-Cosmos, a framework that infuses generative World Models (WM) into the MCP ecosystem to enable predictive task automation. By unifying three disparate technologies, namely MCP, World Model, and Agent, we demonstrate that a "Bring Your Own World Model" (BYOWM) strategy allows agents to simulate state transitions and refine plans in a latent space before execution. We conducted experiments using two strategies, namely ReAct and SPIRAL with 2 planning models and 3 representative world models over 20+ MCP-Bench tasks. We observed improvements in Agent's environment interaction KPI such as tool success rate and tool parameter accuracy. The framework also offers new metrics such as Execution Quality to generate new insights about the effectiveness of world models compared to baseline.
Teaching language models to use search tools is not only a question of whether they search, but also of whether they issue good queries. This is especially important in open-domain question answering, where broad or copied queries often waste retrieval budget and derail later reasoning. We propose \Ours, a framework that makes query planning explicit through reusable search skills. At each step, the model first selects a skill, then generates a search or answer action conditioned on the selected skill card. The skill inventory itself is not fixed: SearchSkill maintains an evolving SkillBank, expands or refines it from recurrent failure patterns, and reconstructs affected trajectories before supervised training. The resulting two-stage SFT recipe aligns training with the inference-time protocol of skill selection followed by skill-grounded execution. Across open-source and closed-source models, SearchSkill improves exact match on knowledge-intensive QA benchmarks and yields better retrieval behavior, including fewer copied first queries, more atomic hop-focused queries, and more correct answers within a small search budget. These results suggest that explicit skill-conditioned query planning is a lightweight alternative to treating search as an undifferentiated action.
Automated intrusion-style workflows require LLM agents to reason over partial observations, tool outputs, and executable artifacts under bounded budgets. A single LLM instance often compresses evidence extraction, planning, execution, and validation into one context, which increases the risk of context drift and error propagation. Existing LLM-based multi-agent systems support general collaboration, but they do not explicitly model the role boundaries, artifact provenance, and cost constraints that characterize multi-stage intrusion workflows. This paper presents CAESAR, a coordinated multi-agent framework for controlled analysis of LLM-agent behavior in intrusion-style tasks. CAESAR decomposes the workflow into five typed roles and coordinates them through a bounded round protocol with a persistent knowledge base, a per-round workspace, validator-gated knowledge promotion, and capability-token write isolation. We evaluate CAESAR on 25 CTF tasks across five categories and four LLM backends. Compared with a single-agent baseline under matched budgets and tool access, CAESAR improves task success and reduces performance variance, with larger gains on tasks requiring multi-step exploit composition. A secondary simulated interactional-security study suggests that the role structure can transfer beyond code-native surfaces. The results indicate that role transitions, artifact provenance, and knowledge-promotion events provide useful structural signals for monitoring coordinated LLM-agent behavior beyond individual prompt and output inspection. The dataset, implementation, and evaluation logs are released at https://github.com/Xu-Qiu/CMAS.
While LLM-based agents excel at planning and executing long action sequences, their execution often remains inconsistent across trials, limiting reliability. Consolidating agent consistency requires distilling trial-error trajectories into reusable skills that preserve task-relevant invariants while discarding trajectory-specific noise. However, in multimodal settings, the key challenge is not only that useful invariants are distributed across vision and language information, but that different modalities support different kinds of reusable skill content: while some skills are verbalizable and interpretable, others reside in perceptual evidence beyond text. Text-only skills may lose perceptual cues, whereas storing text and perception naively introduces redundancy and noise. Existing inference-time methods, such as self-consistency, improve reliability through costly multi-sample decoding, while internalization strategies lack a way to separate verbalizable skill content from residual perceptual information. To address this, we introduce Conditional Multimodal Information Bottleneck (CMIB), a method for multimodal skill construction. CMIB begins with a joint bottleneck over multimodal skills and derives an exact sequential decomposition: (1) a text-stage bottleneck distilling interpretable skill cards, and (2) a conditional multimodal bottleneck compressing only residual information in perception that remains predictive beyond text. Unlike naive two-stream formulations, CMIB explicitly conditions the multimodal latent on the text skill, thus structurally reducing cross-modal redundancy and enabling independent control over textual and perceptual compression. We instantiate CMIB with a variational objective that makes its conditional decomposition tractable to optimize, yielding reusable multimodal skills that improve execution stability without incurring multi-sample inference overhead.
Interactive narrative tasks require LLMs to sustain a coherent, evolving story while adapting to a user over multiple turns. However, suitable benchmarks for this setting are limited: existing evaluations often focus on static prompts, isolated story generations, or post-hoc ratings, and therefore miss whether models can jointly manage story generation, long-context state and pacing, character simulation, empathic personalization, and story-grounded artifacts. We introduce NARRA-Gym, an executable evaluation environment that turns a sparse emotional seed into a complete interactive story episode and logs the full model-in-the-loop trajectory, including story construction, memory updates, planning, pacing interventions, and optional artifact synthesis. We evaluate nine frontier LLMs using a controlled LLM-as-judge sweep over eight benchmark personas and a human evaluation in which participants rate customized model outputs. Our results show substantial variation across models, personas, and evaluation dimensions: models that produce fluent stories can still fail on robustness, user experience, or resistance-sensitive personalization. These findings suggest that interactive narrative offers a useful benchmark for evaluating long-horizon, user-adaptive LLM behavior beyond isolated story quality.
Explicit planning is a critical capability for LLM-based agents solving complex data-centric tasks, which require precise tool calling over external data sources. Existing strategies fall into two paradigms based on planning horizon: (1) full-horizon (FH), which generates a complete plan before execution, and (2) single-step horizon (SH), which interleaves each action (tool call) with incremental reasoning and observation. While step-by-step execution is a common default under the assumption that eager execution monitoring is necessary for adaptability, we revisit this assumption for well-defined data-centric tasks. Our controlled empirical study isolates planning horizon as the key architectural feature and systematically analyzes the effects of topological complexity and tool robustness on both paradigms. Our experiments across Knowledge Base Question Answering and Multi-hop QA show that FH planning with lazy replanning achieves accuracy parity with SH across varying depths, breadths, and robustness levels, while using 2-3x fewer tokens. These findings suggest that for well-defined data-centric tasks, eager step-wise monitoring is often unnecessary, and full-horizon planning with on-demand replanning can offer a more efficient default.
We present a hierarchical language-driven framework for robotic task and motion planning to improve natural, intuitive human-robot interaction in service and assistance scenarios. The proposed system employs two large language model (LLM) modules: a high-level planning agent and a low-level spatial reasoning sub-module. The primary agent processes natural language commands and generates action sequences using a ReAct-style prompt, interacting with tools for object perception and manipulation (e.g., pick, place, release). For precise spatial placement, such as interpreting "place the mug next to the plate", a separate sub-prompting module handles 3D reasoning based on object geometry and scene layout. The system integrates YOLOX-GDRNet for object detection and pose estimation, along with a motion execution stub. We evaluated the system in 24 test scenarios, ranging from simple spatial commands to high-level instructions and infeasible requests. The system achieved an overall task success rate of 86%.
Robot swarms promise scalable assistance in complex and hazardous environments. Task planning lies at the core of human-swarm collaboration, translating the operator's intent into coordinated swarm actions and helping determine when validation or intervention is required during execution. In long-horizon missions under dynamic scenarios, however, reliable task planning becomes difficult to maintain: emerging events and changing conditions demand continual adaptation, and sustained operator oversight imposes substantial cognitive burden. Existing LLM-based planning tools can support plan generation, yet they remain susceptible to invalid task orderings and infeasible robot actions, resulting in frequent manual adjustment. Here we introduce a neuro-symbolic framework for long-horizon human-swarm collaboration that tightly melds verifiable task planning with context-grounded LLM reasoning. We formalize mission goals and operational rules as temporal logic formulas and admissible task orderings as task automata. Conditioned on these formal constraints and live perceptual context, LLMs generate executable subtask sequences that satisfy mission rules and remain grounded in the current scene. An uncertainty-aware scheduler then assigns subtasks across the heterogeneous swarm to maximize parallelisms while remaining resilient to disruptions. An event-triggered interaction protocol further limits operator involvement to sparse, high-level confirmation and guidance. Deployment on a heterogeneous robotic fleet yields similar results while remaining robust to hardware-specific actuation and communication uncertainties. Together, these results support a formal and scalable paradigm for reliable and low-overhead human-swarm collaboration in dynamic environments
Natural-language interfaces can make swarm robotics more accessible to non-expert operators, but they must translate ambiguous user intent into executable swarm behaviors without unsupported actions, malformed programs, or unsafe plans. This paper presents CommandSwarm, a safety-aware language-to-behavior-tree pipeline for generating XML behavior trees (BTs) from speech or text commands. The system combines multilingual translation, command-level safety filtering, constrained prompting, a LoRA-adapted large language model (LLM), and deterministic parser validation against a whitelist of executable swarm primitives. We evaluate eleven open 6.7B--14B parameter LLMs, all using 4-bit quantization, on representative swarm-control scenarios under zero-shot, one-shot, and two-shot prompting. Falcon3-Instruct-10B and Mistral-7B-v3 are the strongest prompt-engineered candidates, reaching BLEU scores above 0.60 and high syntactic validity in few-shot settings. LoRA adaptation of Falcon3-Instruct-10B on a 2,063-example synthetic instruction--BT corpus improves zero-shot BLEU from 0.267 to 0.663, ROUGE-L from 0.366 to 0.692, and parser-accepted syntactic validity from 0% to 72%. Translation experiments further show that SeamlessM4T v2-large and EuroLLM-9B provide the best quality-latency trade-offs for the multilingual front end. The results indicate that compact, quantized, domain-adapted LLMs can generate useful swarm BTs when embedded in a validated systems pipeline. They also show that parser acceptance and safety filtering remain necessary execution gates; generation quality alone is not sufficient for autonomous deployment.
HTN planning is a variation of classical planning where, instead of searching for a linear sequence of actions, an algorithm decomposes higher-level tasks using a method library until only executable actions remain. On one hand, this allows one to introduce domain knowledge that can speed up the search for a solution through the method library. On the other hand, it creates challenges that go beyond those of classical state-space search. While recent research produced a number of heuristics and novel algorithms that speed up HTN planning, these heuristics are not yet as informative as those available in classical planning algorithms. We investigate whether large language models (LLMs) can generate effective search heuristics for HTN planning, extending the methodology of Corrêa, Pereira, and Seipp (2025) from classical to hierarchical planning. Using the Pytrich planner on six standard total-order HTN benchmark domains, we evaluate heuristics generated by nine LLMs under domain-specific prompting and compare them against the TDG and LMCount domain-independent baselines and the PANDA planner. Our results show that LLM-generated heuristics nearly match the coverage of the best available HTN planner, while substantially reducing search effort on 83% of shared problems.
Tourism is a high-stakes setting for conversational recommender systems (CRS): a plausible-sounding suggestion can waste real money and trip time once a traveler acts on it. Existing CRS benchmarks primarily evaluate systems with a single Recall@k score over entity mentions, and tourism-specific resources add spatial or knowledge-graph context, yet none of them couple multi-turn recommendation with verbatim review-span evidence and rejection recovery. This leaves an evaluation gap for tourism recommendation that is simultaneously trustworthy, verifiable, and adaptive: recommend the right point of interest (POI) for multi-aspect preferences (such as cuisine, price, atmosphere, walking distance), justify each suggestion with verifiable evidence from prior visitors so the traveler can act without trial and error, and recover when the first recommendation is rejected mid-dialogue. We introduce TRACE, where each item is a multi-turn tourism recommendation dialogue with review-span citations and explicit rejection turns: 10,000 dialogues over 2,400 Yelp POIs and 34,208 reviews across eight U.S. cities, paired with 14 retrieval, planning, and LLM baselines, along with 25 metrics organized under Accuracy, Grounding, and Recovery. Across these baselines, TRACE reveals the Three-Competency Gap: LLM Zero-Shot leads in closed-set Recall@1 and rejection recovery but cites less densely than retrievers; non-LLM retrievers achieve surface-verbatim grounding but with low accuracy; Multi-Review Synthesis fails at recovery. The Grounding Score agrees with human citation precision (Spearman rho=+0.80, p<10^-20), and paired t-tests reproduce the per-baseline ranking (p<0.01 on the dominant contrasts). TRACE reframes accountable tourism recommendation as a joint target (right POI, verifiable evidence, adaptive repair) rather than a single-axis leaderboard.
Large language model (LLM)-based agents that reason, plan, and act through tools, memory, and structured interaction are emerging as a promising paradigm for automating complex workflows. Recent systems such as OpenClaw and Claude Code exemplify a broader shift from passive response generation to action-oriented task execution. Yet as agents move toward open-ended, real-world deployment, relying on from-scratch reasoning and low-level tool calls for every task become increasingly inefficient, error-prone, and hard to maintain. This survey examines this challenge through the lens of \emph{agent skills}, which we define as reusable procedural artifacts that coordinate tools, memory, and runtime context under task-specific constraints. Under this view, agents and skills play complementary roles: agents handle high-level reasoning and planning, while skills form the operational layer that enables reliable, reusable, and composable execution. Skills are therefore central to the scalability, robustness, and maintainability of modern agent systems. We organize the literature around four stages of the agent skill lifecycle -- representation, acquisition, retrieval, and evolution -- and review representative methods, ecosystem resources, and application settings across each stage. We conclude by discussing open challenges in quality control, interoperability, safe updating, and long-term capability management. All related resources, including research papers, open-source data, and projects, are collected for the community in \textcolor{blue}{https://github.com/JayLZhou/Awesome-Agent-Skills}.
Large Language Models (LLMs) have demonstrated remarkable capabilities in orchestrating tools for reasoning tasks. However, existing methods rely on a step-wise paradigm that lacks a global perspective, which causes error accumulation over long horizons and restricts generalization to unseen tools. To overcome these limitations, we propose Tools as Continuous Flow for Evolving Agentic Reasoning (FlowAgent), which reconceptualizes tool chaining as continuous trajectory generation within a semantic space. To systematically evaluate this paradigm, we introduce the first plan-level closed-loop benchmark dedicated to plan-level agentic reasoning in dynamic real-world environments. Specifically, the proposed FlowAgent leverages conditional flow matching to generate continuous latent trajectories, providing a global planning perspective to ensure coherent and robust tool execution. Theoretically, we establish formal bounds on utility convergence and prove that our continuous formulation fundamentally guarantees robust generalization and error attenuation. Empirical evaluations show that FlowAgent achieves superior robustness and adaptability in long-horizon reasoning tasks.
Large Language Models (LLMs) have become increasingly capable as tool-using agents, with benchmarks spanning diverse general agentic tasks. Yet rigorous evaluation of scientific tool use remains limited. In chemistry, recent agents can plan syntheses and invoke domain-specific tools, but evaluations often rely on curated demonstrations, expert assessment, or LLM-as-judge scoring rather than exact, judge-free ground truth. We address this gap with chemical procurement cost estimation, a practical task in which an agent must ground chemical identities, retrieve supplier quotes, select valid purchasable packs, normalize quantities, and compute cost from a reaction description. We introduce ChemCost, a benchmark of 1,427 evaluable reactions grounded to a frozen pricing snapshot covering 2,261 chemicals and 230,775 supplier quotes, supporting scalar scoring and stage-level diagnosis of grounding, retrieval, procurement, and arithmetic failures. To evaluate robustness, we further construct controlled noise-injected views that perturb chemical aliases, quantity expressions, missing fields, and input formatting. Experiments with frontier, open-weight, and chemistry-specialized LLM agents show that tool access is necessary but insufficient for solving the task. The strongest agents reach only 50.6% accuracy within 25% relative error on clean inputs and degrade substantially with realistic noise. Stage-level analysis further shows that failures arise from brittle parsing, ineffective evidence integration, invalid pack selection, and non-convergent tool use.
Beyond training-time optimization, scaling test-time computation has emerged as a key paradigm to extend the reasoning capabilities of Large Language Models (LLMs). However, most existing methods adopt a rigid Planning-before-Trial (PbT) policy, which inefficiently allocates test-time compute by incurring planning overhead even on directly solvable problems. We propose Planning-after-Trial (PaT), an adaptive policy for code generation that invokes a planner only upon verification failure. This adaptive policy naturally enables a heterogeneous model configuration: a cost-efficient model handles generation attempts, while a powerful model is reserved for targeted planning interventions. Empirically, across multiple benchmarks and model families, our approach significantly advances the cost-performance Pareto frontier. Notably, our heterogeneous configuration achieves performance comparable to a large homogeneous model while reducing inference cost by approximately 69\%.
Large language model (LLM) applications are increasingly executed as heterogeneous multi-stage workflows rather than isolated inference calls. In these workflow directed acyclic graphs (DAGs), scheduling decisions affect not only the currently ready stage, but also the execution state inherited by downstream stages, including model residency, parent-output locality, prefix reuse, and future device reachability. Existing serving and DAG-scheduling policies mainly optimize immediate queue state, placement cost, or reuse signals in isolation, which can fragment useful state and increase end-to-end latency. We present FATE, a future-state-aware scheduler for heterogeneous LLM workflows. FATE combines a CP-SAT-backed frontier planner, horizon-aware candidate scoring, bounded multi-device shard execution, and state-conditional cost estimation. Rather than solving a monolithic full-DAG problem, FATE repeatedly plans over the current ready frontier and scores assignments by both immediate cost and the downstream state they induce. Across real-DAG and controlled prefix-reuse benchmarks, FATE outperforms practical heuristics, classical DAG scheduling, and proxy adaptations of recent workflow-serving policies. On the real-DAG benchmark, it achieves normalized makespan and normalized P95 latency of 0.675 and 0.677, reducing them by 32.5% and 32.3% over RoundRobin and by 8.9% and 8.8% over the strongest non-FATE baseline. Mechanism analysis and ablations show that these gains arise from jointly preserving multiple dimensions of future execution state rather than prefix reuse alone. These results indicate that future-state preservation should be treated as a first-class scheduling objective for heterogeneous LLM workflow serving.
Autonomous systems that build structures from natural-language instructions need reliable spatial reasoning, yet large language models (LLMs) make systematic coordinate errors when generating three-dimensional block placements. We present a neuro-symbolic pipeline based on \emph{2.5-D decomposition}: the LLM plans in the two-dimensional horizontal plane while a deterministic executor computes all vertical placement from column occupancy, eliminating an entire class of errors. On the Build What I Mean benchmark (160 rounds), GPT-4o-mini with this pipeline achieves 94.6\% mean structural accuracy across 12 independent runs, within 3.0 percentage points of the 97.6\% ceiling imposed by architect-agent errors that no builder-side improvement can address. This outperforms both GPT-4o at 90.3\% and the best competing system at 76.3\%. A controlled ablation confirms that 2.5-D decomposition is the dominant contributor, accounting for 50.7 percentage points of accuracy. The pipeline transfers directly to edge hardware: Nemotron-3 120B running locally on an NVIDIA Jetson Thor AGX matches the cloud result at 94.5\% with no prompt modifications. The underlying principle, removing deterministic dimensions from the LLM's output space, applies to any autonomous construction or assembly task where gravity or other physical constraints fix one or more degrees of freedom. A transfer experiment on 500 IGLU collaborative building tasks confirm the effect generalizes beyond the primary benchmark.
Enterprise interest in multi-agent systems has shifted from generic software agents to large-language-model (LLM) based intelligent agents that plan, use tools, maintain contextual memory, inspect intermediate results, collaborate with other agents, and sometimes act in systems of record. This paper revises the enterprise architecture thesis around a design-first claim: governance is necessary, but it cannot be the primary organizing abstraction. The primary abstraction must be agent design - capability boundaries, autonomy allocation, interaction protocols, tool and data authority, state and memory design, verification design, and human interaction design. We propose CEAD (Capability-Aligned Enterprise Agent Design), a reference architecture for intelligent agents that uses service-oriented architecture (SOA) as an exemplar for contracts, registries, loose coupling, and policy-aware integration, while explicitly rejecting the idea that services are agents. It treats microservices as a cautionary precedent: decomposition without design discipline produces distributed complexity, cost, operational fragility, and agent proliferation. We evaluate CEAD over 10,000 enterprise tasks, comparing five architectures: a prompt-first mono-agent, a role-based micro-agent swarm, SOA-brokered agents, a governance-first but design-poor agent grid, and the proposed CEAD architecture. CEAD achieves 70.6% safe success, versus 45.2% for the mono-agent baseline, 23.1% for the ungoverned micro-agent swarm, 58.8% for SOA-brokered agents, and 50.8% for the control-heavy, design-poor grid. The results support the conclusion that design quality is the first-order enterprise concern; governance, security, policy, audit, and assurance should support and enforce good design rather than substitute for it.
We present a dynamic policy-learning approach that combines generalized planning and hierarchical task decomposition for LLM-based agents. Our method, Hierarchical Component Learning for Generalized Policies (HCL-GP ), learns parameterized policies that generalize across task instances and automatically extracts reusable components from successful executions, organizing them into a component library for compositional policy generation. We address three challenges: (1) learning components through automated decomposition, (2) generalizing components to maximize reuse, and (3) efficient retrieval via semantic search. Evaluated on the AppWorld benchmark, our approach achieves 98.2% accuracy on normal tasks and 97.8% on challenge tasks with unseen applications, improving 15.8 points over static synthesis on challenging scenarios. For open-source models, dynamic reuse enables 62.5% success versus near-zero without reuse. This demonstrates that classical planning concepts can be effectively integrated with LLM agents for improved accuracy and efficiency.
AI agent research spans a wide spectrum: from RL agents that learn from scratch to foundation model agents that leverage pre-trained knowledge, yet no unified benchmark enables fair comparison across these approaches. We present Agentick, a benchmark for sequential decision-making agents designed to evaluate RL, LLM, VLM, hybrid, and human agents on common ground and to power research on the fundamental challenges of sequential decision-making. Agentick provides 37 procedurally generated tasks across six capability categories, four difficulty levels, and five observation modalities, all exposed through a single Gymnasium-compatible interface. The benchmark ships with a Coding API, oracle reference policies for all tasks, pre-built SFT datasets, a composable agent harness, and a live leaderboard. An evaluation spanning 27 configurations and over 90,000 episodes reveals that no single approach dominates: GPT-5 mini leads overall at 0.309 oracle-normalized score while PPO dominates planning and multi-agent tasks; the reasoning harness multiplies LLM performance by 3-10x; and ASCII observations consistently outperform natural language. These findings highlight the substantial room for improvement that remains across all agent paradigms. Agentick's capability-decomposed, multi-modal design provides the empirical infrastructure needed to drive progress toward general autonomous agents, both as an evaluation framework and as a training ground for RL post-training of foundation models in truly sequential environments.
Large language models (LLMs), especially reasoning models, generate extended chain-of-thought (CoT) reasoning that often contains explicit deliberation over future outcomes. Yet whether this deliberation constitutes genuine planning, how it is structured, and what aspects of it drive performance remain poorly understood. In this work, we introduce a new method to characterize LLM planning by extracting and quantifying search trees from reasoning traces in the four-in-a-row board game. By fitting computational models on the extracted search trees, we characterize how plans are structured and how they influence move decisions. We find that LLMs' search is shallower than humans', and that performance is predicted by search breadth rather than depth. Most strikingly, although LLMs expand deep nodes in their traces, their move choices are best explained by a myopic model that ignores those nodes entirely. A causal intervention study where we selectively prune CoT paragraphs further suggests that move selection is driven predominantly by shallow rather than deep nodes. These patterns contrast with human planning, where performance is driven primarily by deep search. Together, our findings reveal a key difference between LLM and human planning: while human expertise is driven by deeper search, LLMs do not act on deep lookahead. This dissociation offers targeted guidance for aligning LLM and human planning. More broadly, our framework provides a generalizable approach for interpreting the structure of LLM planning across strategic domains.