Unverified50% confidenceEventExact time
The paper "Functional Gradient Descent with Adaptive Representations" was accepted at NeurIPS.
1
Sources
50%
Confidence
Long-term
Relevance
9/28/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedNeuroevolution uses evolutionary algorithms instead of gradient descent to optimize the weights or architecture of neural networks, requiring no differentiable loss function.64% similarUnverified梯度下降及其变体（Adam、AdaGrad等）本质上是数值优化算法60% similarUnverifiedThe Abliteration technique identifies and suppresses the direction vector responsible for refusal behavior in the model's residual stream, using orthogonal projection to remove that direction from the weight matrices, thereby eliminating refusal behavior without retraining.60% similarUnverified多级调用图能展示函数的上游调用者和下游被调用者，帮助评估改动的潜在影响范围59% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/956481API
curl https://kongchang.com/api/v1/knowledge/claims/956481MCP
get_claim(id=956481)