A well-tuned kNN router matches or exceeds state-of-the-art learned routers on new standardized benchmarks spanning instruction, QA, reasoning, and the first multi-modal visual routing dataset, due to locality of model performance in embedding space.
arXiv preprint arXiv:2405.18137 (2024)
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
representative citing papers
Smaller LLMs produce functional but limited Python code with variable quantization effects and quality/maintainability concerns that require validation before use.
citing papers explorer
-
Rethinking Predictive Modeling for LLM Routing: When Simple kNN Beats Complex Learned Routers
A well-tuned kNN router matches or exceeds state-of-the-art learned routers on new standardized benchmarks spanning instruction, QA, reasoning, and the first multi-modal visual routing dataset, due to locality of model performance in embedding space.
-
Precision or Peril: A PoC of Python Code Quality from Quantized Large Language Models
Smaller LLMs produce functional but limited Python code with variable quantization effects and quality/maintainability concerns that require validation before use.