AI RESEARCH

Reliability-Aware Adaptive Self-Consistency for Efficient Sampling in LLM Reasoning

arXiv CS.LG

ArXi:2601.02970v2 Announce Type: replace-cross Self-Consistency improves reasoning reliability through multi-sample aggregation, but incurs substantial inference cost. Adaptive self-consistency methods mitigate this issue by adjusting the sampling budget; however, they rely on count-based stopping rules that treat all responses equally, often leading to unnecessary sampling.