Unverified50% confidenceFactExact time
The kNN method based on bge-base embeddings (predicting using the outcomes of the 50 most similar historical cases) showed flat performance across all quintiles, with no discriminative power.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/12/2026
First Seen
Valid until: 1/10/2027
Sources
Related Entities
Related Claims
Unverifiedk=5时可以以95%置信度区分成功率0.9和0.7的模型，k=10时能识别更细微的差异64% similarUnverifiedAblation experiments show that when the KL retention coefficient drops from 0.1 to 0.01, the correction rate falls from 53.3% to 35.0%, indicating that KL anchoring also contributes to supporting correction effectiveness.62% similarUnverified报告同时给出中位数KLD和99.9%分位KLD两个数值，前者反映典型情况，后者捕捉极端长尾错误58% similarUnverifiedIn a counterfactual analysis of 500 divergence positions (10,000 MCTS iterations each, rollout depth 20, repeated three times), among the 105 positions where Jev went face while the neural network developed the board, the deep MCTS preferred Jev in 92 cases, preferred the neural network in 3, and produced 10 ties.57% similarUnverifiedUnder the zero-shot setting on the TurnBench development set, FD-VAD achieves a maximum EOT recall of 0.853 under the constraint of false positive rate FP≤0.10.56% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1005470API
curl https://kongchang.com/api/v1/knowledge/claims/1005470MCP
get_claim(id=1005470)