paper-with-me

홈 › Papers

Breaking Model Lock-in: Cost-Efficient Zero-Shot LLM Routing via a Universal Latent Space

2026-01-09 · Cheng Yan, Wuyang Zhang, Zhiyuan Ning, Fan Xu, Ziyang Tao, Lu Zhang, Bing Yin, Yanyong Zhang arxiv

The rapid proliferation of Large Language Models (LLMs) has led to a fragmented and inefficient ecosystem, a state of ``model lock-in'' where seamlessly integrating novel models remains a significant bottleneck. Current routing frameworks require exhaustive, costly retraining, hindering scalability and adaptability. We introduce ZeroRouter, a new paradigm for LLM routing that breaks this lock-in. Our approach is founded on a universal latent space, a model-agnostic representation of query difficulty that fundamentally decouples the characterization of a query from the profiling of a model. This allows for zero-shot onboarding of new models without full-scale retraining. ZeroRouter features a context-aware predictor that maps queries to this universal space and a dual-mode optimizer that balances accuracy, cost, and latency. Our framework consistently outperforms all baselines, delivering higher accuracy at lower cost and latency.

📄 PDF Abstract BibTeX arXiv:2601.06220

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Multi-Agent Routing as Set-Valued Prediction: A WildChat Benchmark and Cost-Aware Evaluation

2026-06-27 · Ananto Nayan Bala, Faisal Muhammad Shah arxiv

Tool and agent routing from natural-language prompts is naturally a set-valued prediction problem: a single query may require multiple agents, while over-selection increases execution cost. The benchmark introduced here …

Parametric Memory Decoding for Zero-Shot Routing in LoRA-Based External Parametric Memory

2026-07-05 · Fengxian Ji, Zhuohan Xie, Jingpu Yang, Fan Zhang 외 arxiv

With the rise of parametric memory, LoRA-based External Parametric Memory (EPM) has emerged as a modular solution, but existing routing methods often introduce additional training, deployment, and maintenance overhead. T…

Dynamic sparsity in tree-structured feed-forward layers at scale

2026-03-18 · Reza Sedghi, Robin Schiewer, Anand Subramoney, David Kappel arxiv

At typical context lengths, the feed-forward MLP block accounts for a large share of a transformer's compute budget, motivating sparse alternatives to dense MLP blocks. We study sparse, tree-structured feed-forward layer…

Question Answering

Prompt Learning for Generalized Vehicle Routing

2024-05-20 · Fei Liu, Xi Lin, Weiduo Liao, Zhenkun Wang 외

Neural combinatorial optimization (NCO) is a promising learning-based approach to solving various vehicle routing problems without much manual algorithm design. However, the current NCO methods mainly focus on the in-dis…

Combinatorial OptimizationPrompt LearningZero-shot Generalization

Zero-Shot Confidence Estimation for Small LLMs: When Supervised Baselines Aren't Worth Training

2026-05-04 · Luong N. Nguyen arxiv

How reliably can a small language model estimate its own correctness? The answer determines whether local-to-cloud routing-escalating queries a cheap local model cannot handle-can work without supervised training data. A…