New paper proposes aspect-wise pairwise tournaments for test-time selection in LLMs
An arXiv paper introduces an aspect-wise pairwise tournament method for test-time selection, aiming to improve on best-of-n sampling for scaling test-time compute in LLM reasoning.
Read the full story at arXiv cs.AI ↗
Timeline · 1 report
- 2026-11-10 05:00 · arXiv cs.AI
Aspect-wise Pairwise Tournaments for Test-Time Selection