Unverified70% confidenceFactTime unknown
The Self-Consistency method, proposed by Xuezhi Wang and others, improved accuracy by more than 10% on mathematical reasoning benchmarks such as GSM8K.
2
Sources
70%
Confidence
Long-term
Relevance
6/1/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedOn the GSM8K mathematical reasoning benchmark, Self-Consistency can improve accuracy by approximately 17 percentage points compared to single greedy decoding.78% similarUnverifiedSelf-Consistency通过多次采样推理路径后投票取多数答案，可进一步提升准确率约5-10个百分点73% similarUnverifiedThe paper reports that hybrid reasoning agents achieve accuracy about 10 percentage points higher than conventional models on multiple test sets.71% similarUnverifiedSelf-Consistency can bring improvements of 5–20 percentage points on mathematical reasoning benchmarks such as GSM8K.69% similarUnverified在GSM8K小学数学应用题基准上，前沿模型的准确率已超过90%69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/28808API
curl https://kongchang.com/api/v1/knowledge/claims/28808MCP
get_claim(id=28808)