Unverified50% confidenceFactExact time
The paper arXiv:2610.09044v1 theoretically characterizes how much improvement weak-to-strong generalization can bring, under what conditions it occurs, and when it disappears
1
Sources
50%
Confidence
Long-term
Relevance
10/8/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe paper points out that the root cause of long-context degradation lies not entirely in context length, but in the number of effective distractors it contains.66% similarUnverified现有的泛化理论大多假设性能随规模平滑变化，无法预测或解释涌现能力这种不连续的跃升66% similarUnverifiedAn arXiv paper (arXiv:2610.07228v1) analyzes the degree of inefficiency of natural gradient descent (NGD), providing a complete picture ranging from exact optimality to unbounded growth with dimensionality.66% similarUnverified对高基数类别特征进行独热编码会导致特征空间急剧膨胀，造成维度灾难，使欧氏距离在高维稀疏空间中区分能力急剧下降64% similarUnverifiedThe vaguer the prompt, the more the model output collapses toward high-probability regions, and the stronger the consistency.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/986974API
curl https://kongchang.com/api/v1/knowledge/claims/986974MCP
get_claim(id=986974)