Scout Before You Route: Predicting Task Success and Token Count from a Small LLM Prefill
cs.AI, cs.LG
Submitted: 2026-11-06
Updated: 2026-11-06
Terminology
Sources
- SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language Model
- Models Under SCOPE: Scalable and Controllable Routing via Pre-hoc Reasoning
- GraphRouter: A Graph-based Router for LLM Selections
- Omni-MATH: A Universal Olympiad Level Mathematic Benchmark For Large Language Models
- The Llama 3 Herd of Models
- Jina Embeddings 2: 8192-Token General-Purpose Text Embeddings for Long Documents
- Measuring Coding Challenge Competence With APPS
- RouterBench: A Benchmark for Multi-LLM Routing System
- LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code
- BIG-Bench Extra Hard
- Most of the LLM Routing Gap Is Task Type
- LLMRouterBench: A Massive Benchmark and Unified Framework for LLM Routing
- Competition-Level Code Generation with AlphaCode
- LLMs Encode Their Failures: Predicting Success from Pre-Generation Activations
- SuperGPQA: Scaling LLM Evaluation across 285 Graduate Disciplines
- Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAs
- RouteLLM: Learning to Route LLMs with Preference Data
- gpt-oss-120b & gpt-oss-20b Model Card
- Route to Reason: Adaptive Routing for LLM and Reasoning Strategy Selection
- Mind the Delay: Latency-aware Query Routing for Heterogeneous LLM pools
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents