Unverified50% confidenceFactExact time
LeanGraphSearch 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.
1
Sources
50%
Confidence
Long-term
Relevance
10/16/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedIn 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 v2.77% similarUnverified对多数应用场景而言，提示词优化的性价比高于模型微调71% similarUnverifiedFor making generated results look good, adversarial loss is often more effective than pixel loss—the former encourages results that look correct, while the latter only pursues numerical closeness.67% similarUnverifiedEnd-to-end task completion rate is very likely to become a more practically valuable model evaluation metric than benchmark scores.67% 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66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1042039API
curl https://kongchang.com/api/v1/knowledge/claims/1042039MCP
get_claim(id=1042039)