Unverified50% confidenceFactExact time
Experiments found that when only a limited number of questions are available, shallow questions often achieve higher task performance than deep questions, even when the total number of questions is the same.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
11/5/2026
First Seen
Valid until: 2/3/2027
Sources
Related Entities
Related Claims
UnverifiedPractical implication: in test-time ensembles, when questions are scarce one should prioritize expanding the question set rather than blindly increasing the number of generations; when questions are abundant, the improvement from the number of generations encounters an identification ceiling.75% similarUnverifiedExperiments on general tasks show that the hierarchical question construction method improves interpretability while maintaining task performance.75% similarUnverifiedThe study observed an inherent difficulty gap across different tasks, with LLMs performing notably weaker on certain types of reasoning tasks.74% similarUnverifiedMechanism three: when the number of questions is sufficiently large, the minimax excess risk reaches the identification lower bound of N^{-β}, which cannot be broken even with infinitely many questions, and this lower bound is independent of the number of questions.73% similarUnverifiedThe benefit of a cascade architecture comes from the long-tail distribution of task difficulty: the vast majority of requests fall into simple categories, so reserving expensive models for the few hard samples lowers average cost with almost no change in quality.73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1138220API
curl https://kongchang.com/api/v1/knowledge/claims/1138220MCP
get_claim(id=1138220)