Maestro: Joint Graph & Config Optimization for Reliable AI Agents

2025-09-04 · Wenxiao Wang, Priyatham Kattakinda, Soheil Feizi arxiv

Building reliable LLM agents requires decisions at two levels: the graph (which modules exist and how information flows) and the configuration of each node (models, prompts, tools, control knobs). Most existing optimizers tune configurations while holding the graph fixed, leaving structural failure modes unaddressed. We introduce Maestro, a framework-agnostic holistic optimizer for LLM agents that jointly searches over graphs and configurations to maximize agent quality, subject to explicit rollout/token budgets. Beyond numeric metrics, Maestro leverages reflective textual feedback from traces to prioritize edits, improving sample efficiency and targeting specific failure modes. On the IFBench and HotpotQA benchmarks, Maestro consistently surpasses leading prompt optimizers--MIPROv2, GEPA, and GEPA+Merge--by an average of 12%, 4.9%, and 4.86%, respectively; even when restricted to prompt-only optimization, it still leads by 9.65%, 2.37%, and 2.41%. Maestro achieves these results with far fewer rollouts than GEPA. We further show large gains on two applications (interviewer & RAG agents), highlighting that joint graph & configuration search addresses structural failure modes that prompt tuning alone cannot fix.

📄 PDF Abstract BibTeX arXiv:2509.04642

Code (0)

등록된 구현이 없습니다.

MAESTRO: Multi-Agent Evaluation Suite for Testing, Reliability, and Observability

2026-01-01 · Tie Ma, Yixi Chen, Vaastav Anand, Alessandro Cornacchia 외 arxiv

We present MAESTRO, an evaluation suite for the testing, reliability, and observability of LLM-based MAS. MAESTRO standardizes MAS configuration and execution through a unified interface, supports integrating both native…

Maestro-U: Leveraging joint speech-text representation learning for zero supervised speech ASR

2022-10-18 · Zhehuai Chen, Ankur Bapna, Andrew Rosenberg, Yu Zhang 외

Training state-of-the-art Automated Speech Recognition (ASR) models typically requires a substantial amount of transcribed speech. In this work, we demonstrate that a modality-matched joint speech and text model can be l…

Representation Learning speech-recognition Speech Recognition Transfer Learning

Understanding Reuse, Performance, and Hardware Cost of DNN Dataflows: A Data-Centric Approach Using MAESTRO

2018-05-04 · Hyoukjun Kwon, Prasanth Chatarasi, Michael Pellauer, Angshuman Parashar 외

The data partitioning and scheduling strategies used by DNN accelerators to leverage reuse and perform staging are known as dataflow, and they directly impact the performance and energy efficiency of DNN accelerator desi…

Scheduling valid

Orchestrating Spatial Semantics via a Zone-Graph Paradigm for Intricate Indoor Scene Generation

2026-05-04 · Meisheng Zhang, Shizhao Sun, Yang Zhao, Ziyuan Liu 외 arxiv

Autonomous 3D indoor scene synthesis breaks down in non-convex rooms with tightly coupled spatial constraints. Data-driven generators lack topological priors for long-horizon planning, while iterative agents fragment sem…

Indoor Scene Synthesis Scene Generation

MAESTRO: Open-Ended Environment Design for Multi-Agent Reinforcement Learning

2023-03-06 · Mikayel Samvelyan, Akbir Khan, Michael Dennis, Minqi Jiang 외

Open-ended learning methods that automatically generate a curriculum of increasingly challenging tasks serve as a promising avenue toward generally capable reinforcement learning agents. Existing methods adapt curricula …

continuous-control Continuous Control Multi-agent Reinforcement Learning reinforcement-learning+2