Unverified50% confidenceBenchmarkExact time
Benchmark tests show that when the system contains long-running transactions holding more than 64 subtransactions, the performance of other concurrent queries exhibits a cliff-like drop, and the degradation is non-linear, erupting suddenly after a critical threshold is crossed.
1
Sources
50%
Confidence
Long-term
Relevance
9/12/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedIn long-context capability tests, models processing inputs of tens of thousands of tokens may exhibit the 'lost in the middle' phenomenon.71% similarUnverifiedOn update-fact tasks, embedding retrieval methods exhibited unpredictable collapses, with success rates fluctuating between 0.30 and 0.95 and a maximum gap of 0.45 across different random seeds.70% similarUnverified增加子线程数量不等于增加智力或能力，当数量超过阈值（如20个甚至100个）会出现重复搜索、相互锚定等问题导致效率下降68% similarUnverifiedIn document chunking strategies, chunks that are too long reduce retrieval precision, while chunks that are too short result in loss of context.68% similarUnverifiedDrift detection involves a trade-off: setting the threshold too sensitively leads to frequent false alarms and unnecessary retraining, while setting it too loosely may allow model performance to continuously degrade without being detected.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/914610API
curl https://kongchang.com/api/v1/knowledge/claims/914610MCP
get_claim(id=914610)