Unverified50% confidenceBenchmarkExact time
In end-to-end evaluations at {8, 16, 32} reflection rounds, LeanGraphSearch improves the pass rate by up to 40% over an ordinary prover and by up to 10% over LeanSearch v2.
1
Sources
50%
Confidence
Long-term
Relevance
10/16/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedLeanGraphSearch has a better efficiency ratio of accuracy to cost in end-to-end evaluations, indicating that the pass rate improvement does not come only from more rounds or greater computation, and that part of the gain comes from retrieval quality itself.77% similarUnverifiedPagedAttention可将GPU显存利用率从通常的20%-40%提升至接近理论上限，吞吐量相比原生HuggingFace推理提升数倍至十余倍66% similarUnverified在多跳推理任务上 GraphRAG 的准确率比传统 RAG 高出 40% 以上，但索引构建时间增加 3-5 倍65% similarUnverifiedIn the article, under DeepEval's default mode, 7 responses had their pass/fail conclusion flip at least once across three re-runs; after manually fixing evaluation_steps, the number of flips dropped to 3, and the agreement rate rose from 31/40 to 35/40.63% similarUnverified实验显示模型对序列首尾信息的召回率比中间位置高出20%～40%63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1042032API
curl https://kongchang.com/api/v1/knowledge/claims/1042032MCP
get_claim(id=1042032)