Unverified50% confidenceBenchmarkExact time
At a 6k budget, the best planner leads by only 0.40 percentage points; by a 9k budget, the planner actually falls behind.
1
Sources
50%
Confidence
Long-term
Relevance
9/24/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedWhen the budget was concentrated on the reflection/control slot, the success rate recovered to 0.761 even with only half of the split budget.69% similarUnverifiedReducing the LoRA weight (typically from the default 1.0 to 0.6–0.8) combined with 6–8 inference steps is a compromise that finds a better balance between efficiency and quality.61% similarUnverifiedFor the 7B model, accuracy with the solver was 75.31% versus 72.02% for direct calculation, a paired improvement of +3.29 percentage points, with a 95% cluster interval of [-3.49, 10.38] crossing zero, making the advantage non-significant.61% similarUnverified在代码补全、函数改写等结构化任务上，中低端模型与顶级模型的输出质量差距通常在5%以内60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/946188API
curl https://kongchang.com/api/v1/knowledge/claims/946188MCP
get_claim(id=946188)