task dag construction

To address the ambiguity and redundancy in DAG serialization arising from non-unique topological orders, this paper proposes a reversible generative framework based on an unambiguous context-free graph grammar: it uniquely maps each DAG to a deterministic sequence of production rules, eliminating reliance on topological sorting while ensuring representation compactness, uniqueness, and invertibility. Methodologically, we formulate DAG generation as an unambiguous syntactic derivation process—the first such formulation—and introduce a normalized traversal strategy coupled with a rule-serialization mechanism. The framework enables lossless DAG compression, differentiable graph generation, continuous latent-space learning for attribute prediction, and efficient Bayesian optimization over structured data. It combines theoretical rigor—guaranteeing bijective mapping—with interpretable, syntax-driven generation.

ROS 2’s publish-subscribe model lacks native support for enforcing priority and data-dependency constraints in directed acyclic graph (DAG)–structured tasks, resulting in out-of-order callback execution, inconsistent multi-input matching policies, and DAG semantics sustained solely through ad hoc programming conventions—rendering systems prone to instability and crashes. To address this, we propose the Function-as-Subtask (FasS) API: a declarative interface that explicitly models data flow via function parameters and return values, thereby enforcing DAG structure at the API level and eliminating reliance on developer discipline. We implement a native DAG-aware scheduler in Rust and design a system integration layer compatible with Linux’s sched_ext subsystem. Experimental evaluation demonstrates that FasS guarantees semantic fidelity while delivering a production-ready, real-time–capable DAG scheduling infrastructure.

Long-horizon collaborative tasks for dual robotic arms face challenges including complex spatiotemporal dependencies among subtasks, difficulty in dynamic action allocation, and limited expressiveness of linear programming formulations. This paper proposes the first LLM-driven DAG-structured task decomposition framework, which automatically parses high-level instructions into directed acyclic graphs (DAGs) encoding dependency constraints, and integrates environment perception to enable real-time, dynamic action allocation and parallel adaptive execution across both arms. The method breaks away from predefined operational paradigms, supporting end-to-end, interpretable, and generalizable collaborative planning. Evaluated on the Dual-Arm Kitchen benchmark, it achieves a 52.8% efficiency gain over single-arm systems, improves success rate by 48% and reduces LLM query count by 84.1% compared to conventional dual-arm planners, significantly enhancing robustness and scalability in complex scenarios.

"This study addresses the challenge of transforming natural language routing rules, authored by business administrators, into executable workflow diagrams for enterprise contact centers. The project employs a neural-symbolic decomposition approach, utilizing a compact intermediate representation and a deterministic compiler to minimize the need for direct graph construction by large language models. A key innovation is the integration of a learned registry selection front-end, which enhances the generation of relevant vocabulary, significantly improving the quality and efficiency of Directed Acyclic Graph (DAG) workflow generation. Implemented across four distinct models, the system achieved an effectiveness score of approximately 89%, with condition accuracy around 90% and JSON format correctness rates of 99%-100%. Additionally, the token count per rule was reduced by half compared to monolithic prompting methods."

This study addresses the brittleness and computational redundancy inherent in static planning for deep research agents by proposing DAGent, a novel framework that introduces a "post-evaluation incremental growth" mechanism to dynamically expand a directed acyclic task graph based on confidence estimates. By integrating hierarchical context management with topology-conditioned reinforcement learning (DAGRPO), optimized via structure-compliant regularization, the framework enhances multi-agent collaborative execution. Experimental results demonstrate that DAGent surpasses state-of-the-art open-source baselines across multiple benchmarks, significantly improving accuracy while substantially reducing both token consumption and reasoning steps.

Existing large language model agents tackling multi-step tasks often rely on costly recomputation or task-specific fine-tuning, resulting in poor generalization and limited reusability of intermediate results. This work proposes the Atomic Task Graph (ATG) framework, which— for the first time—explicitly models task decomposition and execution dependencies using a unified directed acyclic graph. During planning, ATG recursively decomposes high-level tasks; during execution, it enables parallel scheduling and local backtracking for error recovery. Notably, ATG operates effectively across diverse tasks without any training, achieving significant performance gains over strong baselines on three interactive benchmarks using only lightweight 7B–8B parameter models, while simultaneously improving both task success rates and execution efficiency.

This work addresses the limitation of existing research agents that oversimplify complex scientific projects into single tasks, resulting in ambiguous task boundaries, disorganized execution, and missing deliverables—challenges that hinder long-horizon, multi-objective, and dependency-sensitive research planning. To overcome this, the authors propose a graph-guided, project-level planning approach that explicitly decomposes a research project into executable task compositions with clearly attributed contributions and explicit dependencies, leveraging an innovative atomic representation and a directed provenance graph. A lightweight Bernoulli block model optimizes task selection, generating standardized task contracts that specify objectives, dependencies, and constraints, enabling seamless decoupled integration with arbitrary executors. Evaluated on ten scientific benchmarks, the method achieves an average quality score of 7.15, significantly outperforming baselines (4.58 and 5.31), and when integrated with AutoResearchClaw, boosts downstream task accuracy from 0.536 to 0.759.

Current prompt graphs lack a clear definition and standardized terminology, resulting in conceptual ambiguity in practice. This work addresses this gap by proposing a formal definition of prompt graph engineering through conceptual analysis, gray literature review, and systematic categorization. It identifies prompt graphs as first-class, executable, and improvable engineering artifacts and establishes four necessary constitutive conditions along with inclusion and exclusion criteria for operational validation. The proposed definition demonstrates consistent applicability across six major frameworks—including LangGraph and DSPy—thereby offering the field its first operational framework and shared vocabulary. Building on this foundation, the paper outlines a future research agenda structured around four key design tensions inherent to prompt graph development.

This work proposes the first arbitrarily scalable and automatically verifiable task-graph benchmark designed to evaluate language agents’ ability to retain, update, combine, and discard contextual information during complex reasoning. The benchmark constructs task graphs from natural language questions paired with executable Python solvers, modeling tasks through typed intermediate states such as scalars and lists. It enables flexible control over task length, dependency structure, distractors, and value types. Experimental results reveal that while Qwen3.5-27B excels on isolated tasks, its accuracy drops by up to 33.3% on complex tasks involving branching dependencies, effectively exposing a critical bottleneck in current agents’ context management capabilities.

This study addresses the prohibitive computational burden of combinatorial search in exact DAG learning, which arises from the exponential growth of candidate parent sets. To overcome this challenge, we propose DECO, a nonparametric mixture framework that pioneers directional evidence extraction to construct admissible parent sets, thereby substantially reducing the search space while preserving all plausible edge orientations without requiring prespecified parametric models. Theoretically, we prove that DECO achieves an exponential reduction in the search space. Extensive experiments on benchmark Bayesian networks and synthetic datasets demonstrate that the proposed method significantly lowers computational complexity while maintaining highly competitive performance in structure recovery accuracy and Structural Hamming Distance (SHD).