Neural Bandit Based Optimal LLM Selection for a Pipeline of Subtasks

发布

2026年04月23日

采集 2026年04月23日 00:00

行业动态 6.0 分 — Neural bandit approach to LLM routing for multi-step tasks. Practical for LLM pipeline optimization.

原文： arXiv

评分 6.0 · 来源： · 发布于

评分依据：Neural bandit approach to LLM routing for multi-step tasks. Practical for LLM pipeline optimization.

BatchLLM: Optimizing Large Batched LLM Inference with Global Prefix Sharing and Throughput-oriented Token Batching

Foundational Design Principles and Patterns for Building Robust and Adaptive GenAI-Native Systems