aiqbench analysis
MoSAIC learns compact priority functions for DAG task scheduling by fitting them to optimal CP-SAT schedules with Bayesian Optimization, with an LLM assisting the reasoning about graph structure. The reported gains over HEFT on PolyBench and MachSuite come from arXiv-listed experiments on synthetic and benchmark graphs, so the practical value depends on how well the learned heuristics transfer to production HLS and embedded flows. Readers should check the evaluation setup and the runtime overhead claims before treating this as a general scheduling improvement.
MoSAIC: Learning DAG Scheduling Heuristics from Optimal Schedules through Motif-Aware Recovery and LLM-Guided Reasoning
arXiv:2610.15205v1 Announce Type: cross Abstract: Efficient task scheduling in heterogeneous computing systems remains a fundamental challenge in embedded and high-level synthesis flows. This paper presents MoSAIC (Motif-aware Scheduling with AI-driven Cognition), a learning-based framework that bridges the gap between exact and heuristic scheduling approaches for DAG-based applications. The system uses the reasoning capabilities of large language models (LLMs) to learn the relationships between DAG structures and scheduling priority functions. The framework first leverages the OR-Tools CP-SAT solver [4] to generate optimal schedules, which are then used to supervise the learning of compact priority-based scheduling functions through Bayesian Optimization. These learned heuristics capture relationships between scheduling decisions and structural graph features such as rank, depth, and fanout, enabling the construction of a reusable knowledge base across diverse DAG instances. Experimental results across synthetic graph families and real-world benchmarks, including PolyBench and MachSuite, show that the learned heuristics consistently improve over classical approaches such as HEFT [2], while maintaining low runtime overhead. The results indicate that combining optimal supervision with learning-based generalization provides a practical approach for scalable and adaptive task scheduling.
Abstract:Efficient task scheduling in heterogeneous computing systems remains a fundamental challenge in embedded and high-level synthesis flows. This paper presents MoSAIC (Motif-aware Scheduling with AI-driven Cognition), a learning-based framework that bridges the gap between exact and heuristic scheduling approaches for DAG-based applications. The system uses the reasoning capabilities of large language models (LLMs) to learn the relationships between DAG structures and scheduling priority functions. The framework first leverages the OR-Tools CP-SAT solver [4] to generate optimal schedules, which are then used to supervise the learning of compact priority-based scheduling functions through Bayesian Optimization. These learned heuristics capture relationships between scheduling decisions and structural graph features such as rank, depth, and fanout, enabling the construction of a reusable knowledge base across diverse DAG instances. Experimental results across synthetic graph families and real-world benchmarks, including PolyBench and MachSuite, show that the learned heuristics consistently improve over classical approaches such as HEFT [2], while maintaining low runtime overhead. The results indicate that combining optimal supervision with learning-based generalization provides a practical approach for scalable and adaptive task scheduling.
Hardware Architecture (cs.AR); Information Retrieval (cs.IR); Machine Learning (cs.LG)
arXiv:2610.15205 [cs.AR]
(or arXiv:2610.15205v1 [cs.AR] for this version)
https://doi.org/10.48550/arXiv.2610.15205 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 275 words
Related coverage
Vertical Advantage: Transforming Industries with Lakebase and Agentic AI
Databricks · Aug 27, 2026
From Atoms to Entropy: Optimal Noise Allocation for Diffusion Training in the Convex Regime
arXiv · Jul 24, 2026
Implicit Q-learning-bootstrapped ant colony optimization for maritime moving-target observation scheduling with agile satellites
arXiv · Aug 27, 2026
Optimizing Denoising Trajectories in dLLMs: A Lightweight Evolutionary Heuristic Approach
arXiv · Sep 23, 2026