Unverified50% confidenceFactExact time
对抗样本概念由Szegedy等人于2013年在论文《Intriguing properties of neural networks》中正式提出
1
Sources
50%
Confidence
Long-term
Relevance
7/7/2026
First Seen
Sources
Anti-Surveillance Fashion: How Adversarial Examples Can Make Cameras Fail to Recognize You
hackernewshackernews7/5/2026
Related Claims
VerifiedMechanistic interpretability is driven primarily by researchers from institutions such as Anthropic and DeepMind, dedicated to reverse-engineering the algorithmic structures inside neural networks.71% similarUnverifiedBengio等人于2003年提出神经概率语言模型，将离散的n-gram统计模型升维为连续向量空间中的神经网络67% similarUnverifiedThe idea of embedding vectors can be traced back to Word2Vec proposed by Mikolov et al. in 2013, whose core assumption is the distributional hypothesis.66% similarUnverified可解释性研究的核心目标是揭示神经网络模型尤其是大型语言模型内部的工作机制与决策依据65% similarUnverifiedYoshua Bengio等人强调现有神经网络擅长捕捉统计相关性，但缺乏对因果结构的显式建模能力64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/204508API
curl https://kongchang.com/api/v1/knowledge/claims/204508MCP
get_claim(id=204508)