SCX Router: Streaming Zero-Shot Model Selection with a Decoder-KV Classifier and a Real-World Task Ontology
cs.AI, cs.CL
Submitted: 2026-09-02
Updated: 2026-09-02
Code: https://github.com/Knowledgator/GLiClass
Terminology
Sources
- AutoMix: Automatically Mixing Language Models
- FrugalGPT: How to Use Large Language Models While Reducing Cost and Improving Performance
- A Unified Approach to Routing and Cascading for LLMs
- DeBERTa: Decoding-enhanced BERT with Disentangled Attention
- RouterBench: A Benchmark for Multi-LLM Routing System
- SWE-bench: Can Language Models Resolve Real-World GitHub Issues?
- BIG-Bench Extra Hard
- Focal Loss for Dense Object Detection
- AgentBench: Evaluating LLMs as Agents
- GAIA: a benchmark for General AI Assistants
- RouteLLM: Learning to Route LLMs with Preference Data
- Qwen3 Technical Report
- MuSR: Testing the Limits of Chain-of-thought with Multistep Soft Reasoning
- GLiClass: Generalist Lightweight Model for Sequence Classification Tasks
- Replacing Judges with Juries: Evaluating LLM Generations with a Panel of Diverse Models
- LiveBench: A Challenging, Contamination-Limited LLM Benchmark
- OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments
- GLiNER: Generalist Model for Named Entity Recognition using Bidirectional Transformer
- Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena
- Instruction-Following Evaluation for Large Language Models
Related papers
- 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
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems
- VSAL: A Vision Solver with Adaptive Layouts for Graph Property Detection