Unverified50% confidenceFactExact time
The double descent phenomenon and the grokking phenomenon show that training dynamics are more complex than traditional learning theory predicts.
1
Sources
50%
Confidence
Long-term
Relevance
10/20/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedAs the scale of pretrained models increases, many previously effective explicit knowledge injection techniques have gradually become ineffective or even produced negative effects.70% similarUnverifiedThe VICO paper points out that when model capability improves, fixed training tasks deviate from its learning frontier, causing the learning signal to collapse.70% similarUnverifiedTraining stability is crucial in reinforcement learning, because policy updates, sparse rewards, and environment randomness all cause curve fluctuations, and some self-distillation methods may exhibit a sudden drop in reward, entropy collapse (the policy converging prematurely to a single behavior), or abnormal gradients midway through training.69% similarUnverifiedThe loss surface of multi-layer neural networks is filled with local minima and saddle points, and training requires optimization techniques such as momentum, adaptive learning rates, regularization, and batch normalization.69% similarUnverifiedGAIL借鉴GAN思想用对抗训练隐式恢复专家行为分布，理论基础是将模仿学习转化为占用度量匹配问题，但受GAN模式崩塌问题影响训练稳定性69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1075538API
curl https://kongchang.com/api/v1/knowledge/claims/1075538MCP
get_claim(id=1075538)