Unverified50% confidenceBenchmarkExact time
When responses were frozen at the initial policy, increasing prompt breadth pulled accuracy down from 21.16% to 19.05%.
1
Sources
50%
Confidence
Long-term
Relevance
9/23/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedUnder the condition of refreshing responses at every update, increasing prompt breadth improved accuracy from 23.61% to 25.57%.73% similarUnverifiedLID accuracy at switching points is typically 10–20 percentage points lower than LID accuracy at non-switching points.73% similarUnverifiedSimply rearranging the order of few-shot examples can cause accuracy to fluctuate by more than 20 percentage points.73% similarVerifiedIf the accuracy per step is 95%, after 20 steps the overall accuracy will drop to approximately 36%.71% similarUnverifiedAfter introducing an explicit resolver to handle information supersession relationships, stale-memory failures dropped to zero and overall accuracy jumped from 33% to 78%.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/942910API
curl https://kongchang.com/api/v1/knowledge/claims/942910MCP
get_claim(id=942910)