paper-with-me

Papers

SCOPE-Router: Cost-Aware Open-Set VLM Routing for Execution-Oriented Tasks

2026-08-12 · Tao Yu, Yifei Qu, Zhiqing Cui, Pengfei Zhou, Zhongtian Luo, Yujia Yang, Shenghua Chai, Haopeng Jin, Zhenghao Zhang, Xinming Wang, Hongzhu Yi, Wangbo Zhao, Zhenglin Wan, Yan Huang, Yeshani, Jinwen Luo, Yang You arxiv

Model routing aims to select the most suitable model from a candidate pool for each query, balancing quality and cost. Existing VLM routing research is limited to traditional VQA evaluation, lacks systematic calibration optimization for open-set scenarios, and employs training objectives that dilute multi-positive signals via softmax normalization without incorporating cost. We address these limitations with three contributions: (1)VLM-ExecRouterBench, the first execution-oriented VLM routing benchmark covering Code, Agentic, and Search domains with 11 candidate models spanning nearly two orders of magnitude in pricing; (2)SCOPE-Router, a dual-tower router that matches queries to model behavior profiles constructed via hybrid calibration (random/diagnostic/diversity sampling), enabling new models to join routing without retraining; (3)CRM+RCCR, an architecture-agnostic cost-aware objective that encodes cost preference into continuous relevance targets through per-pair independent scoring, eliminating multi-positive dilution while regularizing queries with similar routing preferences to be closer in the routing space. Empirically, SCOPE-Router achieves the best Rank Score on all three benchmarks, surpassing the runner-up by 1.84 points under OOD settings and by 6.75 points under doubly OOD open-set evaluation. When applied to four diverse routers, CRM+RCCR improves Rank Score by 1.25--6.21 points.

📄 PDF Abstract BibTeX arXiv:2608.12127

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

CARROT: A Cost Aware Rate Optimal Router

2025-02-05 · Seamus Somerstep, Felipe Maia Polo, Allysson Flavio Melo de Oliveira, Prattyush Mangal 외

With the rapid growth in the number of Large Language Models (LLMs), there has been a recent interest in LLM routing, or directing queries to the cheapest LLM that can deliver a suitable response. We conduct a minimax an…

Computational Efficiency

xRouter: Training Cost-Aware LLMs Orchestration System via Reinforcement Learning

2025-10-09 · Cheng Qian, Zuxin Liu, Shirley Kokane, Akshara Prabhakar 외 arxiv

Modern LLM deployments confront a widening cost-performance spectrum: premium models deliver strong reasoning but are expensive, while lightweight models are economical yet brittle on complex tasks. Static escalation rul…

Reinforcement Learning

From Myopic Selection to Long-Horizon Awareness: Sequential LLM Routing for Multi-Turn Dialogue

2026-04-14 · Jiarui Zhang, Xiangyu Liu, Yong Hu, Chaoyue Niu 외 arxiv

Multi-turn dialogue is the predominant form of interaction with large language models (LLMs). While LLM routing is effective in single-turn settings, existing methods fail to maximize cumulative performance in multi-turn…

Route to Rome Attack: Directing LLM Routers to Expensive Models via Adversarial Suffix Optimization

2026-04-16 · Haochun Tang, Yuliang Yan, Jiahua Lu, Huaxiao Liu 외 arxiv

Cost-aware routing dynamically dispatches user queries to models of varying capability to balance performance and inference cost. However, the routing strategy introduces a new security concern that adversaries may manip…

RouteJudge: An Open Platform for Reproducible and Preference-Aware LLM Routing

2026-06-17 · Guannan Lai, Haoran Hu, Han-Jia Ye arxiv

We present RouteJudge, an online pairwise preference evaluation framework for LLM routing systems, with a public platform available at https://routejudge.cn. Different from model-level response evaluation, RouteJudge foc…