Unverified50% confidenceFactExact time
MAPLE's surface-level metrics measure the recovery and breakage of answers, where recovery means a question originally answered incorrectly is answered correctly after KEE intervention, and breakage means a question originally answered correctly is instead answered incorrectly.
1
Sources
50%
Confidence
Long-term
Relevance
10/15/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe core idea for fixing silent failures is: recovery and reporting are two separate jobs and should not be merged; wrappers should retry as before but record the attempt count and mark the span before returning a final failure; a step cap should hand back a separate termination reason; a narrow catch should record the exception before recovering; the parser should run in strict mode; and when falling back, the model that actually answered should be recorded.67% similarUnverifiedEvidence recall is the proportion of the source text fragments on which the final answer depends that were actually read in the trajectory, and it can be used to distinguish retrieval failures from transcription failures.65% similarUnverified原帖作者提出一个恢复检查思路：在一次运行中于某个 step 边界处中断；检查恢复时究竟是哪个档案与哪个模型版本一起被加载；如果出现错配，恢复流程应当暴露出来，而不是静默继续。63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1038559API
curl https://kongchang.com/api/v1/knowledge/claims/1038559MCP
get_claim(id=1038559)