{"slug": "learning-compositional-meta-routing-for-agentic-workflows-an-executable", "title": "Learning Compositional Meta-Routing for Agentic Workflows: An Executable Benchmark", "summary": "Researchers introduced an executable benchmark and a budget-aware meta-router for agentic workflows, achieving 100% success on held-out test tasks versus 93.5% for static workflows, with 43% lower cost. The benchmark includes 216 training, 72 development, 108 held-out test, and 108 locked lexical-shift challenge tasks across data analysis, frozen-corpus research, and document processing. On the untouched challenge split, learned success fell to 75.9%, trailing static routing at 93.5%, identifying lexical generalization as the principal limitation.", "body_md": "arXiv:2608.00106v1 Announce Type: new\nAbstract: Agentic systems must decide not only what answer to produce, but which reasoning and execution operations should precede it. A controller may answer directly, decompose a request, retrieve evidence, execute code, delegate to a specialist, or verify an intermediate result. Existing routing work largely selects model endpoints, retrieval depth, or tools in isolation. We introduce an executable benchmark and a budget-aware meta-router that composes heterogeneous operations from raw task text. The benchmark contains 216 training, 72 development, 108 held-out test, and 108 locked lexical-shift challenge tasks across data analysis, frozen-corpus research, and document processing. Outcomes are machine checked after operations execute. Independent regularized logistic heads predict operation probabilities from word and character features, are temperature-scaled on development data, and are greedily composed under route-cost and action-count budgets. On the held-out test, the learned policy achieves 100% success versus 93.5% for strong static and fixed workflows, with 43% lower cost than the static policy; a matched learned one-shot router reaches 56.5%. On the untouched challenge split, learned success falls to 75.9% and trails static routing at 93.5%, while remaining 49% cheaper and exceeding one-shot routing by 34.3 points. The gap identifies lexical generalization, rather than route execution, as the principal limitation. These results establish a reproducible testbed and a bounded proof of concept, not evidence of live-LLM performance.", "url": "https://wpnews.pro/news/learning-compositional-meta-routing-for-agentic-workflows-an-executable", "canonical_source": "https://arxiv.org/abs/2608.00106", "published_at": "2026-08-04 04:00:00+00:00", "updated_at": "2026-08-04 04:33:34.487704+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "ai-agents"], "entities": ["arXiv"], "alternates": {"html": "https://wpnews.pro/news/learning-compositional-meta-routing-for-agentic-workflows-an-executable", "markdown": "https://wpnews.pro/news/learning-compositional-meta-routing-for-agentic-workflows-an-executable.md", "text": "https://wpnews.pro/news/learning-compositional-meta-routing-for-agentic-workflows-an-executable.txt", "jsonld": "https://wpnews.pro/news/learning-compositional-meta-routing-for-agentic-workflows-an-executable.jsonld"}}